app.py

import json
import os
import requests

import gradio as gr
import numpy as np


base_url = "https://api.sandbox.deepgram.com/nlu"
token_str = os.environ['DG_TOKEN']
def tts_fn(text, speed, pitch_steps, variability):
    texts = [text]
    response = requests.post(
        f'{base_url}', 
        files=[('texts', ('texts', json.dumps(texts), 'application/json'))], 
        params={'synthesize': 'true', 'speed': speed, 'pitch_steps': int(pitch_steps), 'variability': variability},
        headers={
            'Authorization': f'Token {token_str}'
        },
    ).json()
    sample_rate = int(response['results'][0]['sample_rate'])
    audio = np.array(response['results'][0]['audio'], dtype=np.float32)
    return (sample_rate, audio)


app = gr.Blocks()

with app:
    with gr.Tab("TTS MVP"):
        with gr.Row():
            with gr.Column():
                pangram = "The beige hue on the waters of the loch impressed all, including the French queen, before she heard that symphony again, just as young Arthur wanted."
                cherry = "Good evening. I'm a text to speech bot. Please allow me to assist you."
                textbox = gr.TextArea(label="Text", placeholder="Type a sentence here", value=cherry)
                speed = gr.Slider(minimum=0.0, maximum=2.0, value=1.1, step=0.1, label="Speed")
                pitch_steps = gr.Slider(minimum=-24, maximum=24, value=0, step=1, label="Pitch Steps: 12 to an octave")
                variability = gr.Slider(minimum=0.0, maximum=1.0, value=0.7, step=0.1, label="Variability")

            with gr.Column():
                audio_output = gr.Audio(label="Output Audio", elem_id='tts-audio')
                btn = gr.Button("Generate")
                btn.click(tts_fn, inputs=[textbox, speed, pitch_steps, variability], outputs=[audio_output])
app.launch(share=True)