Upload folder using huggingface_hub

.ipynb_checkpoints/Coqui.ai-Copy1-checkpoint.ipynb

@@ -2,8 +2,8 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
-   "id": "14a326bb",
+   "execution_count": 4,
+   "id": "156133fe",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -12,6 +12,7 @@
     "import torch\n",
     "import torch.nn.functional as F\n",
     "from pathlib import Path\n",
+    "import tempfile\n",
     "\n",
     "from TTS.api import TTS\n",
     "from TTS.utils.manage import ModelManager"
@@ -19,8 +20,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
-   "id": "2cfd77d1",
+   "execution_count": 6,
+   "id": "5e5af800",
    "metadata": {
     "scrolled": false
    },
@@ -34,8 +35,8 @@
       " > Check https://choosealicense.com/licenses/mit/ for more info.\n",
       " > Using model: freevc\n",
       " > Loading pretrained speaker encoder model ...\n",
-      "Loaded the voice encoder model on cpu in 0.02 seconds.\n",
-      "Running on local URL:  http://127.0.0.1:7873\n",
+      "Loaded the voice encoder model on cpu in 0.01 seconds.\n",
+      "Running on local URL:  http://127.0.0.1:7863\n",
       "\n",
       "To create a public link, set `share=True` in `launch()`.\n"
      ]
@@ -43,7 +44,7 @@
     {
      "data": {
       "text/html": [
-       "<div><iframe src=\"http://127.0.0.1:7873/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+       "<div><iframe src=\"http://127.0.0.1:7863/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
       ],
       "text/plain": [
        "<IPython.core.display.HTML object>"
@@ -56,7 +57,7 @@
      "data": {
       "text/plain": []
      },
-     "execution_count": 17,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -151,46 +152,985 @@
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
       "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
       " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 3.2769999504089355\n",
-      " > Real-time factor: 0.37722315040572285\n",
-      " > voice_conversion_models/multilingual/vctk/freevc24 is already downloaded.\n",
-      " > Model's license - MIT\n",
-      " > Check https://choosealicense.com/licenses/mit/ for more info.\n",
-      " > Using model: freevc\n",
-      " > Loading pretrained speaker encoder model ...\n",
-      "Loaded the voice encoder model on cpu in 0.02 seconds.\n",
-      "model: voice_conversion_models/multilingual/vctk/freevc24\n",
+      " > Processing time: 3.3410003185272217\n",
+      " > Real-time factor: 0.38459038289093944\n",
+      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 2.9179999828338623\n",
+      " > Real-time factor: 0.3358978221135079\n",
+      " > tts_models/en/ljspeech/tacotron2-DDC_ph is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ljspeech/univnet is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: Tacotron2\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log10\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:True\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:50.0\n",
+      " | > mel_fmax:7600.0\n",
+      " | > pitch_fmin:0.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--tacotron2-DDC_ph\\scale_stats.npy\n",
+      " | > base:10\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Model's reduction rate `r` is set to: 2\n",
+      " > Vocoder Model: univnet\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log10\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:True\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:50.0\n",
+      " | > mel_fmax:7600.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\vocoder_models--en--ljspeech--univnet\\scale_stats.npy\n",
+      " | > base:10\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Generator Model: univnet_generator\n",
+      " > Discriminator Model: univnet_discriminator\n",
+      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
+      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
+      " > Processing time: 3.021000385284424\n",
+      " > Real-time factor: 0.3477544400242312\n",
+      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 2.9099998474121094\n",
+      " > Real-time factor: 0.33497690776101013\n",
+      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 2.933000087738037\n",
+      " > Real-time factor: 0.33762451937136506\n",
+      " > tts_models/en/ljspeech/tacotron2-DDC is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: Tacotron2\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Model's reduction rate `r` is set to: 1\n",
+      " > Vocoder Model: hifigan\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:False\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Generator Model: hifigan_generator\n",
+      " > Discriminator Model: hifigan_discriminator\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Removing weight norm...\n",
+      "model: tts_models/en/ljspeech/tacotron2-DDC\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 4.28600001335144\n",
+      " > Real-time factor: 0.42371906516498953\n",
+      " > tts_models/en/ek1/tacotron2 is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ek1/wavegrad is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: Tacotron2\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log10\n",
+      " | > min_level_db:-10\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:0\n",
+      " | > fft_size:1024\n",
+      " | > power:1.8\n",
+      " | > preemphasis:0.99\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:True\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:10\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Model's reduction rate `r` is set to: 2\n",
+      " > Vocoder Model: wavegrad\n",
+      "model: tts_models/en/ek1/tacotron2\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
+      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
+      " > Processing time: 224.84099984169006\n",
+      " > Real-time factor: 29.51038122922182\n",
+      " > tts_models/en/ek1/tacotron2 is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ek1/wavegrad is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: Tacotron2\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log10\n",
+      " | > min_level_db:-10\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:0\n",
+      " | > fft_size:1024\n",
+      " | > power:1.8\n",
+      " | > preemphasis:0.99\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:True\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:10\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Model's reduction rate `r` is set to: 2\n",
+      " > Vocoder Model: wavegrad\n",
+      "model: tts_models/en/ek1/tacotron2\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
+      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
+      " > Processing time: 266.6489999294281\n",
+      " > Real-time factor: 34.99768124073744\n",
+      " > tts_models/en/ljspeech/tacotron2-DDC_ph is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ljspeech/univnet is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: Tacotron2\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log10\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:True\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:50.0\n",
+      " | > mel_fmax:7600.0\n",
+      " | > pitch_fmin:0.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--tacotron2-DDC_ph\\scale_stats.npy\n",
+      " | > base:10\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Model's reduction rate `r` is set to: 2\n",
+      " > Vocoder Model: univnet\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log10\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:True\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:50.0\n",
+      " | > mel_fmax:7600.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\vocoder_models--en--ljspeech--univnet\\scale_stats.npy\n",
+      " | > base:10\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Generator Model: univnet_generator\n",
+      " > Discriminator Model: univnet_discriminator\n",
+      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
+      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
+      " > Processing time: 2.885999917984009\n",
+      " > Real-time factor: 0.3322142195933605\n",
+      " > Downloading model to C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--speedy-speech\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: speedy_speech\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Vocoder Model: hifigan\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:False\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Generator Model: hifigan_generator\n",
+      " > Discriminator Model: hifigan_discriminator\n",
+      "Removing weight norm...\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "model: tts_models/en/ljspeech/speedy-speech\n",
+      "language: \n",
+      "speaker: \n",
+      "Using original voice\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
+      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
+      " > Processing time: 0.9679999351501465\n",
+      " > Real-time factor: 0.11673301633083617\n",
+      "model: tts_models/en/ljspeech/speedy-speech\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 0.9630000591278076\n",
+      " > Real-time factor: 0.11613007144605443\n",
+      " > tts_models/en/ljspeech/tacotron2-DCA is already downloaded.\n",
+      " > Model's license - MPL\n",
+      " > Check https://www.mozilla.org/en-US/MPL/2.0/ for more info.\n",
+      " > vocoder_models/en/ljspeech/multiband-melgan is already downloaded.\n",
+      " > Model's license - MPL\n",
+      " > Check https://www.mozilla.org/en-US/MPL/2.0/ for more info.\n",
+      " > Using model: Tacotron2\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log10\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:True\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:50.0\n",
+      " | > mel_fmax:7600.0\n",
+      " | > pitch_fmin:0.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--tacotron2-DCA\\scale_stats.npy\n",
+      " | > base:10\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Model's reduction rate `r` is set to: 2\n",
+      " > Vocoder Model: multiband_melgan\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log10\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:0\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:True\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:50.0\n",
+      " | > mel_fmax:7600.0\n",
+      " | > pitch_fmin:0.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\vocoder_models--en--ljspeech--multiband-melgan\\scale_stats.npy\n",
+      " | > base:10\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Generator Model: multiband_melgan_generator\n",
+      " > Discriminator Model: melgan_multiscale_discriminator\n",
+      "model: tts_models/en/ljspeech/tacotron2-DCA\n",
+      "language: \n",
+      "speaker: \n",
+      "Using original voice\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
+      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
+      " > Processing time: 2.067000150680542\n",
+      " > Real-time factor: 0.23295588670728015\n",
+      "model: tts_models/en/ljspeech/tacotron2-DCA\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 2.1570000648498535\n",
+      " > Real-time factor: 0.2430990934225715\n",
+      "model: tts_models/en/ljspeech/tacotron2-DCA\n",
       "language: \n",
       "speaker: \n",
       "voice cloning with the voice conversion model\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.8229997158050537\n",
-      " > Real-time factor: 0.3249621185552823\n",
-      "model: voice_conversion_models/multilingual/vctk/freevc24\n",
+      " > Processing time: 2.0920000076293945\n",
+      " > Real-time factor: 0.23577343069302087\n",
+      " > Downloading model to C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--fast_pitch\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: fast_pitch\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Vocoder Model: hifigan\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:False\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Generator Model: hifigan_generator\n",
+      " > Discriminator Model: hifigan_discriminator\n",
+      "Removing weight norm...\n",
+      "model: tts_models/en/ljspeech/fast_pitch\n",
+      "language: \n",
+      "speaker: \n",
+      "Using original voice\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
+      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
+      " > Processing time: 1.8829996585845947\n",
+      " > Real-time factor: 0.19894272496832988\n",
+      "model: tts_models/en/ljspeech/fast_pitch\n",
       "language: \n",
       "speaker: \n",
       "voice cloning with the voice conversion model\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.858999729156494\n",
-      " > Real-time factor: 0.32910616452921765\n",
-      "model: voice_conversion_models/multilingual/vctk/freevc24\n",
+      " > Processing time: 1.8359999656677246\n",
+      " > Real-time factor: 0.19397711228808903\n",
+      "model: tts_models/en/ljspeech/fast_pitch\n",
       "language: \n",
       "speaker: \n",
       "voice cloning with the voice conversion model\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.8419997692108154\n",
-      " > Real-time factor: 0.3271492592669274\n",
+      " > Processing time: 1.8659999370574951\n",
+      " > Real-time factor: 0.19714666998293168\n",
       "model: voice_conversion_models/multilingual/vctk/freevc24\n",
+      "source_wav: C:\\Users\\Torch\\AppData\\Local\\Temp\\gradio\\b6e9c24083a878478ebbecd7bc42e1f631c05df6\\henry5-0-100.wav\n",
+      "target_wav: C:\\Users\\Torch\\AppData\\Local\\Temp\\gradio\\11c82c70d145ea630f81dfa541de52bf615719ae\\yearn_for_time-0-100.wav\n",
+      " > Downloading model to C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--overflow\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: OverFlow\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      " > Vocoder Model: hifigan\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:False\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Generator Model: hifigan_generator\n",
+      " > Discriminator Model: hifigan_discriminator\n",
+      "Removing weight norm...\n",
+      "model: tts_models/en/ljspeech/overflow\n",
+      "language: \n",
+      "speaker: \n",
+      "Using original voice\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 2.4030001163482666\n",
+      " > Real-time factor: 0.26459208495864933\n",
+      "model: tts_models/en/ljspeech/overflow\n",
+      "language: \n",
+      "speaker: \n",
+      "voice cloning with the voice conversion model\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 2.4769999980926514\n",
+      " > Real-time factor: 0.27343925203231617\n",
+      " > Downloading model to C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--neural_hmm\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
+      " > Model's license - apache 2.0\n",
+      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
+      " > Using model: NeuralHMM_TTS\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Vocoder Model: hifigan\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:False\n",
+      " | > trim_db:60\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Generator Model: hifigan_generator\n",
+      " > Discriminator Model: hifigan_discriminator\n",
+      "Removing weight norm...\n",
+      "model: tts_models/en/ljspeech/neural_hmm\n",
+      "language: \n",
+      "speaker: \n",
+      "Using original voice\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 2.3940000534057617\n",
+      " > Real-time factor: 0.27230367477713896\n",
+      "model: tts_models/en/ljspeech/neural_hmm\n",
       "language: \n",
       "speaker: \n",
       "voice cloning with the voice conversion model\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.922999858856201\n",
-      " > Real-time factor: 0.3364733695695124\n"
+      " > Processing time: 2.628000020980835\n",
+      " > Real-time factor: 0.2965699745262212\n",
+      " > Downloading model to C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--vctk--fast_pitch\n",
+      " > Model's license - CC BY-NC-ND 4.0\n",
+      " > Check https://creativecommons.org/licenses/by-nc-nd/4.0/ for more info.\n",
+      " > Using model: fast_pitch\n",
+      " > Setting up Audio Processor...\n",
+      " | > sample_rate:22050\n",
+      " | > resample:False\n",
+      " | > num_mels:80\n",
+      " | > log_func:np.log\n",
+      " | > min_level_db:-100\n",
+      " | > frame_shift_ms:None\n",
+      " | > frame_length_ms:None\n",
+      " | > ref_level_db:20\n",
+      " | > fft_size:1024\n",
+      " | > power:1.5\n",
+      " | > preemphasis:0.0\n",
+      " | > griffin_lim_iters:60\n",
+      " | > signal_norm:False\n",
+      " | > symmetric_norm:True\n",
+      " | > mel_fmin:0\n",
+      " | > mel_fmax:8000.0\n",
+      " | > pitch_fmin:0.0\n",
+      " | > pitch_fmax:640.0\n",
+      " | > spec_gain:1.0\n",
+      " | > stft_pad_mode:reflect\n",
+      " | > max_norm:4.0\n",
+      " | > clip_norm:True\n",
+      " | > do_trim_silence:True\n",
+      " | > trim_db:23\n",
+      " | > do_sound_norm:False\n",
+      " | > do_amp_to_db_linear:True\n",
+      " | > do_amp_to_db_mel:True\n",
+      " | > do_rms_norm:False\n",
+      " | > db_level:None\n",
+      " | > stats_path:None\n",
+      " | > base:2.718281828459045\n",
+      " | > hop_length:256\n",
+      " | > win_length:1024\n",
+      " > Init speaker_embedding layer.\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
+      "language: \n",
+      "speaker: VCTK_p225\n",
+      "Using original voice\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
+      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
+      " > Processing time: 4.122999906539917\n",
+      " > Real-time factor: 0.6120216766695737\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
+      "language: \n",
+      "speaker: VCTK_p227\n",
+      "Using original voice\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
+      " > Processing time: 3.615000009536743\n",
+      " > Real-time factor: 0.5239715910962163\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
+      "language: \n",
+      "speaker: VCTK_p227\n",
+      "voice cloning with the tts\n",
+      " > Text splitted to sentences.\n",
+      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n"
      ]
     },
     {
@@ -210,21 +1150,17 @@
       "    return await future\n",
       "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\anyio\\_backends\\_asyncio.py\", line 807, in run\n",
       "    result = context.run(func, *args)\n",
-      "TypeError: voice_clone() takes 2 positional arguments but 3 were given\n",
-      "Traceback (most recent call last):\n",
-      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\gradio\\routes.py\", line 434, in run_predict\n",
-      "    event_data=event_data,\n",
-      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\gradio\\blocks.py\", line 1324, in process_api\n",
-      "    fn_index, inputs, iterator, request, event_id, event_data\n",
-      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\gradio\\blocks.py\", line 1052, in call_function\n",
-      "    fn, *processed_input, limiter=self.limiter\n",
-      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\anyio\\to_thread.py\", line 34, in run_sync\n",
-      "    func, *args, cancellable=cancellable, limiter=limiter\n",
-      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\anyio\\_backends\\_asyncio.py\", line 877, in run_sync_in_worker_thread\n",
-      "    return await future\n",
-      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\anyio\\_backends\\_asyncio.py\", line 807, in run\n",
-      "    result = context.run(func, *args)\n",
-      "TypeError: voice_clone() takes 2 positional arguments but 3 were given\n"
+      "  File \"<ipython-input-6-20fd07aa6e62>\", line 65, in text_to_speech\n",
+      "    speech = tts_model.tts(text, language=language, speaker_wav=target_wav)\n",
+      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\TTS\\api.py\", line 548, in tts\n",
+      "    **kwargs,\n",
+      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\TTS\\utils\\synthesizer.py\", line 340, in tts\n",
+      "    speaker_embedding = self.tts_model.speaker_manager.compute_embedding_from_clip(speaker_wav)\n",
+      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\TTS\\tts\\utils\\managers.py\", line 365, in compute_embedding_from_clip\n",
+      "    embedding = _compute(wav_file)\n",
+      "  File \"C:\\ProgramData\\Anaconda3\\lib\\site-packages\\TTS\\tts\\utils\\managers.py\", line 342, in _compute\n",
+      "    waveform = self.encoder_ap.load_wav(wav_file, sr=self.encoder_ap.sample_rate)\n",
+      "AttributeError: 'NoneType' object has no attribute 'load_wav'\n"
      ]
     }
    ],
@@ -269,10 +1205,10 @@
     "def voice_clone(source_wav, target_wav):\n",
     "    print(f'model: {VC_MODEL.model_name}\\nsource_wav: {source_wav}\\ntarget_wav: {target_wav}')\n",
     "    sample_rate = VC_MODEL.voice_converter.output_sample_rate\n",
-    "    if vc_model is None or source_wav is None or target_wav is None:\n",
+    "    if source_wav is None or target_wav is None:\n",
     "        return (sample_rate, np.zeros(0).astype(np.int16))\n",
     "\n",
-    "    speech = vc_model.voice_conversion(source_wav=source_wav, target_wav=target_wav)\n",
+    "    speech = VC_MODEL.voice_conversion(source_wav=source_wav, target_wav=target_wav)\n",
     "    speech = (np.array(speech) * INT16MAX).astype(np.int16)\n",
     "    return (sample_rate, speech)\n",
     "\n",
@@ -291,12 +1227,17 @@
     "    if use_original_voice:\n",
     "        print('Using original voice')\n",
     "        speech = tts_model.tts(text, language=language, speaker=speaker)       \n",
-    "    elif tts_model.synthesizer.tts_model.speaker_manager:\n",
+    "    elif tts_model.synthesizer.tts_model.speaker_manager and tts_model.synthesizer.tts_model.speaker_manager.encoder_ap:\n",
     "        print('voice cloning with the tts')\n",
     "        speech = tts_model.tts(text, language=language, speaker_wav=target_wav)\n",
     "    else:\n",
     "        print('voice cloning with the voice conversion model')\n",
-    "        speech = tts_model.tts_with_vc(text, language=language, speaker_wav=target_wav)\n",
+    "#         speech = tts_model.tts_with_vc(text, language=language, speaker_wav=target_wav)\n",
+    "        with tempfile.NamedTemporaryFile(suffix=\".wav\", delete=False) as fp:\n",
+    "            # Lazy code... save it to a temp file to resample it while reading it for VC\n",
+    "            tts_model.tts_to_file(text, language=language, speaker=speaker, file_path=fp.name)\n",
+    "        speech = VC_MODEL.voice_conversion(source_wav=fp.name, target_wav=target_wav)\n",
+    "        \n",
     "\n",
     "    speech = (np.array(speech) * INT16MAX).astype(np.int16)\n",
     "    return (sample_rate, speech)\n",
@@ -304,7 +1245,7 @@
     "\n",
     "with gr.Blocks() as demo:\n",
     "    tts_model = gr.State(None)\n",
-    "    vc_model = gr.State(None)\n",
+    "#     vc_model = gr.State(None)\n",
     "    def activate(*args):\n",
     "        return gr.update(interactive=True) if len(args) == 1 else [gr.update(interactive=True)] * len(args)\n",
     "    def deactivate(*args):\n",

Coqui.ai-Copy1.ipynb

@@ -3,7 +3,7 @@
   {
    "cell_type": "code",
    "execution_count": 4,
-   "id": "80ca0f5c",
+   "id": "9b361b8e",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -20,8 +20,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
-   "id": "4ad622cb",
+   "execution_count": 7,
+   "id": "38f89cca",
    "metadata": {
     "scrolled": false
    },
@@ -36,7 +36,7 @@
       " > Using model: freevc\n",
       " > Loading pretrained speaker encoder model ...\n",
       "Loaded the voice encoder model on cpu in 0.01 seconds.\n",
-      "Running on local URL:  http://127.0.0.1:7863\n",
+      "Running on local URL:  http://127.0.0.1:7864\n",
       "\n",
       "To create a public link, set `share=True` in `launch()`.\n"
      ]
@@ -44,7 +44,7 @@
     {
      "data": {
       "text/html": [
-       "<div><iframe src=\"http://127.0.0.1:7863/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+       "<div><iframe src=\"http://127.0.0.1:7864/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
       ],
       "text/plain": [
        "<IPython.core.display.HTML object>"
@@ -57,7 +57,7 @@
      "data": {
       "text/plain": []
      },
-     "execution_count": 6,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -65,504 +65,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " > tts_models/en/ljspeech/tacotron2-DDC_ph is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ljspeech/univnet is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: Tacotron2\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:50.0\n",
-      " | > mel_fmax:7600.0\n",
-      " | > pitch_fmin:0.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--tacotron2-DDC_ph\\scale_stats.npy\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Model's reduction rate `r` is set to: 2\n",
-      " > Vocoder Model: univnet\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:50.0\n",
-      " | > mel_fmax:7600.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\vocoder_models--en--ljspeech--univnet\\scale_stats.npy\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Generator Model: univnet_generator\n",
-      " > Discriminator Model: univnet_discriminator\n",
-      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
-      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 3.3410003185272217\n",
-      " > Real-time factor: 0.38459038289093944\n",
-      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.9179999828338623\n",
-      " > Real-time factor: 0.3358978221135079\n",
-      " > tts_models/en/ljspeech/tacotron2-DDC_ph is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ljspeech/univnet is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: Tacotron2\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:50.0\n",
-      " | > mel_fmax:7600.0\n",
-      " | > pitch_fmin:0.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--tacotron2-DDC_ph\\scale_stats.npy\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Model's reduction rate `r` is set to: 2\n",
-      " > Vocoder Model: univnet\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:50.0\n",
-      " | > mel_fmax:7600.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\vocoder_models--en--ljspeech--univnet\\scale_stats.npy\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Generator Model: univnet_generator\n",
-      " > Discriminator Model: univnet_discriminator\n",
-      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
-      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 3.021000385284424\n",
-      " > Real-time factor: 0.3477544400242312\n",
-      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.9099998474121094\n",
-      " > Real-time factor: 0.33497690776101013\n",
-      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.933000087738037\n",
-      " > Real-time factor: 0.33762451937136506\n",
-      " > tts_models/en/ljspeech/tacotron2-DDC is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: Tacotron2\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:False\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:2.718281828459045\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Model's reduction rate `r` is set to: 1\n",
-      " > Vocoder Model: hifigan\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:False\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:False\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:2.718281828459045\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Generator Model: hifigan_generator\n",
-      " > Discriminator Model: hifigan_discriminator\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Removing weight norm...\n",
-      "model: tts_models/en/ljspeech/tacotron2-DDC\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 4.28600001335144\n",
-      " > Real-time factor: 0.42371906516498953\n",
-      " > tts_models/en/ek1/tacotron2 is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ek1/wavegrad is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: Tacotron2\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-10\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:0\n",
-      " | > fft_size:1024\n",
-      " | > power:1.8\n",
-      " | > preemphasis:0.99\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Model's reduction rate `r` is set to: 2\n",
-      " > Vocoder Model: wavegrad\n",
-      "model: tts_models/en/ek1/tacotron2\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
-      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 224.84099984169006\n",
-      " > Real-time factor: 29.51038122922182\n",
-      " > tts_models/en/ek1/tacotron2 is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ek1/wavegrad is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: Tacotron2\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-10\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:0\n",
-      " | > fft_size:1024\n",
-      " | > power:1.8\n",
-      " | > preemphasis:0.99\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Model's reduction rate `r` is set to: 2\n",
-      " > Vocoder Model: wavegrad\n",
-      "model: tts_models/en/ek1/tacotron2\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
-      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 266.6489999294281\n",
-      " > Real-time factor: 34.99768124073744\n",
-      " > tts_models/en/ljspeech/tacotron2-DDC_ph is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ljspeech/univnet is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: Tacotron2\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:50.0\n",
-      " | > mel_fmax:7600.0\n",
-      " | > pitch_fmin:0.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--tacotron2-DDC_ph\\scale_stats.npy\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Model's reduction rate `r` is set to: 2\n",
-      " > Vocoder Model: univnet\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:50.0\n",
-      " | > mel_fmax:7600.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\vocoder_models--en--ljspeech--univnet\\scale_stats.npy\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Generator Model: univnet_generator\n",
-      " > Discriminator Model: univnet_discriminator\n",
-      "model: tts_models/en/ljspeech/tacotron2-DDC_ph\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
-      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 2.885999917984009\n",
-      " > Real-time factor: 0.3322142195933605\n",
-      " > Downloading model to C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--speedy-speech\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: speedy_speech\n",
+      " > tts_models/en/vctk/fast_pitch is already downloaded.\n",
+      " > Model's license - CC BY-NC-ND 4.0\n",
+      " > Check https://creativecommons.org/licenses/by-nc-nd/4.0/ for more info.\n",
+      " > Using model: fast_pitch\n",
       " > Setting up Audio Processor...\n",
       " | > sample_rate:22050\n",
       " | > resample:False\n",
@@ -580,49 +86,14 @@
       " | > symmetric_norm:True\n",
       " | > mel_fmin:0\n",
       " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
+      " | > pitch_fmin:0.0\n",
       " | > pitch_fmax:640.0\n",
       " | > spec_gain:1.0\n",
       " | > stft_pad_mode:reflect\n",
       " | > max_norm:4.0\n",
       " | > clip_norm:True\n",
       " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:2.718281828459045\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Vocoder Model: hifigan\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:False\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:False\n",
-      " | > trim_db:60\n",
+      " | > trim_db:23\n",
       " | > do_sound_norm:False\n",
       " | > do_amp_to_db_linear:True\n",
       " | > do_amp_to_db_mel:True\n",
@@ -632,347 +103,65 @@
       " | > base:2.718281828459045\n",
       " | > hop_length:256\n",
       " | > win_length:1024\n",
-      " > Generator Model: hifigan_generator\n",
-      " > Discriminator Model: hifigan_discriminator\n",
-      "Removing weight norm...\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "model: tts_models/en/ljspeech/speedy-speech\n",
+      " > Init speaker_embedding layer.\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
       "language: \n",
-      "speaker: \n",
+      "speaker: VCTK_p225\n",
       "Using original voice\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
       "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
       " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 0.9679999351501465\n",
-      " > Real-time factor: 0.11673301633083617\n",
-      "model: tts_models/en/ljspeech/speedy-speech\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 0.9630000591278076\n",
-      " > Real-time factor: 0.11613007144605443\n",
-      " > tts_models/en/ljspeech/tacotron2-DCA is already downloaded.\n",
-      " > Model's license - MPL\n",
-      " > Check https://www.mozilla.org/en-US/MPL/2.0/ for more info.\n",
-      " > vocoder_models/en/ljspeech/multiband-melgan is already downloaded.\n",
-      " > Model's license - MPL\n",
-      " > Check https://www.mozilla.org/en-US/MPL/2.0/ for more info.\n",
-      " > Using model: Tacotron2\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:50.0\n",
-      " | > mel_fmax:7600.0\n",
-      " | > pitch_fmin:0.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--tacotron2-DCA\\scale_stats.npy\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Model's reduction rate `r` is set to: 2\n",
-      " > Vocoder Model: multiband_melgan\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log10\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:0\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:True\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:50.0\n",
-      " | > mel_fmax:7600.0\n",
-      " | > pitch_fmin:0.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:C:\\Users\\Torch\\AppData\\Local\\tts\\vocoder_models--en--ljspeech--multiband-melgan\\scale_stats.npy\n",
-      " | > base:10\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Generator Model: multiband_melgan_generator\n",
-      " > Discriminator Model: melgan_multiscale_discriminator\n",
-      "model: tts_models/en/ljspeech/tacotron2-DCA\n",
+      " > Processing time: 3.823000192642212\n",
+      " > Real-time factor: 0.5674894593370367\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
       "language: \n",
-      "speaker: \n",
+      "speaker: VCTK_p248\n",
       "Using original voice\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
-      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 2.067000150680542\n",
-      " > Real-time factor: 0.23295588670728015\n",
-      "model: tts_models/en/ljspeech/tacotron2-DCA\n",
-      "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.1570000648498535\n",
-      " > Real-time factor: 0.2430990934225715\n",
-      "model: tts_models/en/ljspeech/tacotron2-DCA\n",
+      " > Processing time: 3.246999979019165\n",
+      " > Real-time factor: 0.5186337327405872\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
       "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
+      "speaker: VCTK_p267\n",
+      "Using original voice\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.0920000076293945\n",
-      " > Real-time factor: 0.23577343069302087\n",
-      " > Downloading model to C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--fast_pitch\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: fast_pitch\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:False\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:2.718281828459045\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Vocoder Model: hifigan\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:False\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:False\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:2.718281828459045\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Generator Model: hifigan_generator\n",
-      " > Discriminator Model: hifigan_discriminator\n",
-      "Removing weight norm...\n",
-      "model: tts_models/en/ljspeech/fast_pitch\n",
+      " > Processing time: 3.6530003547668457\n",
+      " > Real-time factor: 0.5277034710600691\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
       "language: \n",
-      "speaker: \n",
+      "speaker: VCTK_p334\n",
       "Using original voice\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      "ɛvɹiwɛɹ ðə t͡ʃaɪld wɛnt,\n",
-      " [!] Character '͡' not found in the vocabulary. Discarding it.\n",
-      " > Processing time: 1.8829996585845947\n",
-      " > Real-time factor: 0.19894272496832988\n",
-      "model: tts_models/en/ljspeech/fast_pitch\n",
+      " > Processing time: 3.421999931335449\n",
+      " > Real-time factor: 0.5260981320137958\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
       "language: \n",
-      "speaker: \n",
-      "voice cloning with the voice conversion model\n",
+      "speaker: VCTK_p362\n",
+      "Using original voice\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 1.8359999656677246\n",
-      " > Real-time factor: 0.19397711228808903\n",
-      "model: tts_models/en/ljspeech/fast_pitch\n",
+      " > Processing time: 3.9649999141693115\n",
+      " > Real-time factor: 0.545335878913631\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
       "language: \n",
-      "speaker: \n",
+      "speaker: VCTK_p362\n",
       "voice cloning with the voice conversion model\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 1.8659999370574951\n",
-      " > Real-time factor: 0.19714666998293168\n",
-      "model: voice_conversion_models/multilingual/vctk/freevc24\n",
-      "source_wav: C:\\Users\\Torch\\AppData\\Local\\Temp\\gradio\\b6e9c24083a878478ebbecd7bc42e1f631c05df6\\henry5-0-100.wav\n",
-      "target_wav: C:\\Users\\Torch\\AppData\\Local\\Temp\\gradio\\11c82c70d145ea630f81dfa541de52bf615719ae\\yearn_for_time-0-100.wav\n",
-      " > Downloading model to C:\\Users\\Torch\\AppData\\Local\\tts\\tts_models--en--ljspeech--overflow\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > vocoder_models/en/ljspeech/hifigan_v2 is already downloaded.\n",
-      " > Model's license - apache 2.0\n",
-      " > Check https://choosealicense.com/licenses/apache-2.0/ for more info.\n",
-      " > Using model: OverFlow\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:False\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:True\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:2.718281828459045\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " > Vocoder Model: hifigan\n",
-      " > Setting up Audio Processor...\n",
-      " | > sample_rate:22050\n",
-      " | > resample:False\n",
-      " | > num_mels:80\n",
-      " | > log_func:np.log\n",
-      " | > min_level_db:-100\n",
-      " | > frame_shift_ms:None\n",
-      " | > frame_length_ms:None\n",
-      " | > ref_level_db:20\n",
-      " | > fft_size:1024\n",
-      " | > power:1.5\n",
-      " | > preemphasis:0.0\n",
-      " | > griffin_lim_iters:60\n",
-      " | > signal_norm:False\n",
-      " | > symmetric_norm:True\n",
-      " | > mel_fmin:0\n",
-      " | > mel_fmax:8000.0\n",
-      " | > pitch_fmin:1.0\n",
-      " | > pitch_fmax:640.0\n",
-      " | > spec_gain:1.0\n",
-      " | > stft_pad_mode:reflect\n",
-      " | > max_norm:4.0\n",
-      " | > clip_norm:True\n",
-      " | > do_trim_silence:False\n",
-      " | > trim_db:60\n",
-      " | > do_sound_norm:False\n",
-      " | > do_amp_to_db_linear:True\n",
-      " | > do_amp_to_db_mel:True\n",
-      " | > do_rms_norm:False\n",
-      " | > db_level:None\n",
-      " | > stats_path:None\n",
-      " | > base:2.718281828459045\n",
-      " | > hop_length:256\n",
-      " | > win_length:1024\n",
-      " > Generator Model: hifigan_generator\n",
-      " > Discriminator Model: hifigan_discriminator\n",
-      "Removing weight norm...\n",
-      "model: tts_models/en/ljspeech/overflow\n",
-      "language: \n",
-      "speaker: \n",
-      "Using original voice\n",
-      " > Text splitted to sentences.\n",
-      "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.4030001163482666\n",
-      " > Real-time factor: 0.26459208495864933\n",
-      "model: tts_models/en/ljspeech/overflow\n",
+      " > Processing time: 3.944000005722046\n",
+      " > Real-time factor: 0.5424476055774146\n",
+      "model: tts_models/en/vctk/fast_pitch\n",
       "language: \n",
-      "speaker: \n",
+      "speaker: VCTK_p305\n",
       "voice cloning with the voice conversion model\n",
       " > Text splitted to sentences.\n",
       "['Mary had a little lamb,', 'Its fleece was white as snow.', 'Everywhere the child went,', 'The little lamb was sure to go.']\n",
-      " > Processing time: 2.4769999980926514\n",
-      " > Real-time factor: 0.27343925203231617\n"
+      " > Processing time: 3.994999647140503\n",
+      " > Real-time factor: 0.5442476165199195\n"
      ]
     }
    ],
@@ -1039,7 +228,7 @@
     "    if use_original_voice:\n",
     "        print('Using original voice')\n",
     "        speech = tts_model.tts(text, language=language, speaker=speaker)       \n",
-    "    elif tts_model.synthesizer.tts_model.speaker_manager:\n",
+    "    elif tts_model.synthesizer.tts_model.speaker_manager and tts_model.synthesizer.tts_model.speaker_manager.encoder_ap:\n",
     "        print('voice cloning with the tts')\n",
     "        speech = tts_model.tts(text, language=language, speaker_wav=target_wav)\n",
     "    else:\n",

app.py

@@ -71,7 +71,7 @@ def text_to_speech(text, tts_model, language, speaker, target_wav, use_original_
     if use_original_voice:
         print('Using original voice')
         speech = tts_model.tts(text, language=language, speaker=speaker)       
-    elif tts_model.synthesizer.tts_model.speaker_manager:
+    elif tts_model.synthesizer.tts_model.speaker_manager and tts_model.synthesizer.tts_model.speaker_manager.encoder_ap:
         print('voice cloning with the tts')
         speech = tts_model.tts(text, language=language, speaker_wav=target_wav)
     else: