| import torch | |
| from transformers import SpeechT5Processor, SpeechT5ForTextToSpeech | |
| import soundfile as sf | |
| # Initialize the model and processor from Hugging Face | |
| processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts") | |
| model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts") | |
| def text_to_speech(text): | |
| # Process the input text into tokens | |
| inputs = processor(text, return_tensors="pt") | |
| # Generate speech | |
| with torch.no_grad(): | |
| speech = model.generate_speech(inputs.input_ids) | |
| # Save the generated speech as a WAV file | |
| sf.write('output.wav', speech.squeeze().cpu().numpy(), 16000) | |
| return "output.wav" | |