Spaces:
Sleeping
Sleeping
File size: 2,674 Bytes
e2dc557 |
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 |
import gradio as gr
from split_audio.main import AudioSplitter
import numpy as np
import time
import os
import soundfile as sf
import librosa
splitter = AudioSplitter(language="vi")
def split_audio(str_raw, str_trunc, audio_input):
audio_input = audio_input[1]
y_cut = splitter.split_audio(str_raw, str_trunc, audio_input)
intervals = librosa.effects.split(y_cut, top_db=30)
y_cut = np.concatenate([y_cut[start:end] for start, end in intervals])
return (24000, y_cut)
with gr.Blocks() as demo:
# with gr.Row():
# text_input = gr.Textbox(value="Đây là một ví dụ về tổng hợp giọng nói.")
# audio_output = gr.Audio()
# with gr.Row():
# run_button = gr.Button(value="generate voice")
# rtf_log = gr.Number(label="Real Time Factor")
with gr.Row():
with gr.Column():
text_raw = gr.Textbox(value="", label="Text raw", interactive=False, lines=3)
text_cut = gr.Textbox(label="temp", lines=3)
run_button = gr.Button(value="Run")
with gr.Column():
audio_input = gr.Audio(label="Audio raw", interactive=False, )
audio_output = gr.Audio(label="Temp", interactive=False)
# run_button.click(fn=generate_voice, inputs=[text_input], outputs=[audio_output])
# run_button.click(fn=generate_voice, inputs=[text_input], outputs=[audio_output, rtf_log, phonemizer],)
run_button.click(fn=split_audio, inputs=[text_raw, text_cut, audio_input], outputs=[audio_output])
# get all file wavs in "audio_example" folder
audio_files = [f for f in os.listdir("audio_example") if f.endswith(".wav")]
audio_files = audio_files[:30]
# text_files = [f.replace(".wav", ".txt") for f in audio_files]
# audio = [sf.read(os.path.join("audio_example", f)) for f in audio_files]
# text = [open(os.path.join("audio_example", f), "r", encoding="utf-8").read().strip() for f in text_files]
examples_data = []
for wav_file in audio_files:
# waveform, sr = sf.read(os.path.join("audio_example", wav_file))
waveform, sr = librosa.load(os.path.join("audio_example", wav_file), sr=24000)
intervals = librosa.effects.split(waveform, top_db=30)
waveform = np.concatenate([waveform[start:end] for start, end in intervals])
with open(os.path.join("audio_example", wav_file.replace(".wav", ".txt")), "r", encoding="utf-8") as f:
text = f.read().strip()
examples_data.append([text, (sr, waveform)])
gr.Examples(examples=examples_data, inputs=[text_raw, audio_input])
if __name__ == "__main__":
demo.launch(server_port=7860, server_name="0.0.0.0")
|