edit

app.py

@@ -1,11 +1,9 @@
 import gradio as gr
 import numpy as np
 import torch
-import os
 from datasets import load_dataset
 
 from transformers import AutoProcessor, SpeechT5ForTextToSpeech, SpeechT5HifiGan, SpeechT5Processor, pipeline
-os.environ["CUDA_VISIBLE_DEVICES"]="7"
 
 # device = "cuda" if torch.cuda.is_available() else "cpu"
 device = "cpu"
@@ -14,10 +12,6 @@ device = "cpu"
 asr_pipe = pipeline("automatic-speech-recognition", model="openai/whisper-base", device=device)
 
 # load text-to-speech checkpoint and speaker embeddings
-# processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
-# model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts").to(device)
-# vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan").to(device)
-
 processor = SpeechT5Processor.from_pretrained("sanchit-gandhi/speecht5_tts_vox_nl")
 model = SpeechT5ForTextToSpeech.from_pretrained("sanchit-gandhi/speecht5_tts_vox_nl")
 vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan").to(device)