app.py

import gradio as gr
import time
import torch
import scipy
from espnet2.bin.tts_inference import Text2Speech
from espnet2.utils.types import str_or_none

lang = 'English'
tag = 'kan-bayashi/ljspeech_vits' 
vocoder_tag = "none" 

text2speech = Text2Speech.from_pretrained(
    model_tag=str_or_none(tag),
    vocoder_tag=str_or_none(vocoder_tag),
    device="cpu",
    # Only for Tacotron 2 & Transformer
    threshold=0.5,
    # Only for Tacotron 2
    minlenratio=0.0,
    maxlenratio=10.0,
    use_att_constraint=False,
    backward_window=1,
    forward_window=3,
    # Only for FastSpeech & FastSpeech2 & VITS
    speed_control_alpha=1.0,
    # Only for VITS
    noise_scale=0.333,
    noise_scale_dur=0.333,
)


def inference(text):
  with torch.no_grad():
      start = time.time()
      wav = text2speech(text)["wav"]
  scipy.io.wavfile.write("out.wav",text2speech.fs , wav.view(-1).cpu().numpy())
  return  "out.wav"
title = "TTS"
description = "demo for Anime2Sketch. To use it, simply upload your image, or click one of the examples to load them. Read more at the links below."
article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2104.05703'>Adversarial Open Domain Adaption for Sketch-to-Photo Synthesis</a> | <a href='https://github.com/Mukosame/Anime2Sketch'>Github Repo</a></p>"

gr.Interface(
    inference, 
    "text", 
    gr.outputs.Audio(type="file", label="Output"),
    title=title,
    description=description,
    article=article,
    enable_queue=True
    ).launch(debug=True)