File size: 1,419 Bytes
4110922 d33f32d 4110922 6afc51e 4110922 4af5dfa 4110922 4af5dfa 4110922 0a5aead 45fac78 0a5aead 45fac78 0a5aead 4af5dfa 45fac78 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 |
import torch
from TTS.api import TTS
import gradio as gr
import os
import spaces
# Agree to Terms of service
# os.environ["COQUI_TOS_AGREED"] = "1"
def init_TTS():
# Get device
device = "cuda" if torch.cuda.is_available() else "cpu"
# Initialize the TTS model
tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to(device)
return tts
@spaces.GPU
def generate_speech(text):
# Generate speech using the provided text, speaker voice, and language
file_path = "output.wav"
speaker_wav = "/content/speaker.wav"
language = "en"
tts.tts_to_file(text=text,
file_path=file_path,
speaker_wav=speaker_wav,
language=language)
return file_path
def main():
# call init
tts = init_TTS()
# Create the Gradio interface
interface = gr.Interface(
fn=generate_speech,
inputs=[
gr.Textbox(label="Enter your text")
#gr.Textbox(label="Path to target speaker WAV file", value="/content/speaker.wav")
#gr.Dropdown(label="Language", choices=["en"], value="en")
],
outputs="audio",
title="Voice Synthesis with Coqui-XTTS",
description="Synthesize speech using predefined target voice and language."
)
# Launch the interface
interface.launch()
return 0
if __name__ == "__main__":
main()
|