fix on way to apply batched fater-whiper

app.py

@@ -4,7 +4,7 @@ import tempfile
 
 import torch
 import gradio as gr
-from faster_whisper import WhisperModel
+from faster_whisper import BatchedInferencePipeline, WhisperModel
 from pydub import AudioSegment
 from pyannote.audio import Pipeline as DiarizationPipeline
 import opencc
@@ -101,11 +101,12 @@ def get_fwhisper_model(model_id: str, device: str) -> WhisperModel:
     key = (model_id, device)
     if key not in _fwhisper_models:
         compute_type = "float16" if device.startswith("cuda") else "int8"
-        _fwhisper_models[key] = WhisperModel(
+        model = WhisperModel(
             model_id,
             device=device,
             compute_type=compute_type,
         )
+        _fwhisper_models[key] = BatchedInferencePipeline(model=model)
     return _fwhisper_models[key]
 
 def get_sense_model(model_id: str, device_str: str):