File size: 870 Bytes
e7db7ab 03fc43e 8d12a0e 03fc43e e35048f 5b3e648 73316e5 6608fe5 1692178 3a1d118 03fc43e 78e4f6a 6608fe5 16803e6 8255a9c 16803e6 56d59fe ef3739a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 |
import tempfile ,os
import gradio as gr
from transformers import VitsModel, AutoTokenizer,pipeline
import torch
import numpy as np
import torchaudio
model = VitsModel.from_pretrained("SeyedAli/Persian-Speech-synthesis")
tokenizer = AutoTokenizer.from_pretrained("SeyedAli/Persian-Speech-synthesis")
text_input = gr.TextArea(label="متن فارسی",text_align="right",rtl=True,type="text")
audio_output = gr.Audio(label="صوت گفتار فارسی", type="filepath")
def TTS(text):
inputs = tokenizer(text, return_tensors="pt")
with torch.no_grad():
output = model(**inputs).waveform
with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as fp:
torchaudio.save(fp, output, model.config.sampling_rate,format="wav")
return fp.name
iface = gr.Interface(fn=TTS, inputs=text_input, outputs=audio_output)
iface.launch(share=False) |