Elesis's picture
Update app.py
7a5b815
raw
history blame
No virus
944 Bytes
import time
import streamlit as st
import numpy as np
import torch
from espnet2.bin.tts_inference import Text2Speech
from scipy.io.wavfile import write
from PIL import Image
fs, lang = 44100, "Japanese"
model= "./100epoch.pth"
x = "JR東ζ—₯ζœ¬γ‚’γ”εˆ©η”¨γγ γ•γ„γΎγ—γ¦γ€γ‚γ‚ŠγŒγ¨γ†γ”γ–γ„γΎγ™γ€‚"
text2speech = Text2Speech.from_pretrained(
model_file=model,
device="cpu",
speed_control_alpha=1.0,
noise_scale=0.333,
noise_scale_dur=0.333,
)
pause = np.zeros(30000, dtype=np.float32)
st.title("JREast_Auto_announce_TTS")
text = st.text_area(label='γ“γ“γ«γƒ†γ‚­γ‚Ήγƒˆγ‚’ε…₯εŠ› (Input Text)↓', height=100, max_chars=2048)
if st.button("η”ŸζˆοΌˆGenerateοΌ‰"):
with torch.no_grad():
wav = text2speech(text)["wav"]
wav_list = []
wav_list.append(np.concatenate([wav.view(-1).cpu().numpy(), pause]))
final_wav = np.concatenate(wav_list)
st.audio(final_wav, sample_rate=fs)