Spaces:
Runtime error
Runtime error
from transformers import SpeechT5ForTextToSpeech, SpeechT5Processor, SpeechT5HifiGan | |
import torch | |
import gradio as gr | |
# Load the TTS model and processor | |
model_checkpoint = "ejazhabibdar/speecht5_finetuned_voxpopuli_nl" | |
model = SpeechT5ForTextToSpeech.from_pretrained(model_checkpoint) | |
processor = SpeechT5Processor.from_pretrained(model_checkpoint) | |
# Load the vocoder for generating speech | |
vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan") | |
# Set the device | |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu") | |
model.to(device) | |
vocoder.to(device) | |
# Set the model to evaluation mode | |
model.eval() | |
vocoder.eval() | |
# Define the TTS function | |
def text_to_speech(text): | |
# Preprocess the input text | |
inputs = processor(text=text, return_tensors="pt") | |
inputs = {k: v.to(device) for k, v in inputs.items()} | |
# Generate speech | |
with torch.no_grad(): | |
speech = model.generate_speech(inputs["input_ids"], vocoder=vocoder) | |
return speech.tolist() | |
# Create a Gradio interface | |
iface = gr.Interface( | |
fn=text_to_speech, | |
inputs="text", | |
outputs="audio", | |
title="Text-to-Speech", | |
description="Enter the text and listen to the generated speech.", | |
theme="huggingface", | |
) | |
if __name__ == "__main__": | |
iface.launch() | |