Spaces:
Running
on
T4
Running
on
T4
Commit
•
532dc11
1
Parent(s):
23840ce
Update app.py
Browse files
app.py
CHANGED
@@ -1,10 +1,14 @@
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
import os, torch, io
|
3 |
-
os.system('python -m unidic download')
|
|
|
4 |
from melo.api import TTS
|
5 |
speed = 1.0
|
6 |
import tempfile
|
7 |
-
|
|
|
8 |
models = {
|
9 |
'EN': TTS(language='EN', device=device),
|
10 |
'ES': TTS(language='ES', device=device),
|
@@ -14,22 +18,44 @@ models = {
|
|
14 |
'KR': TTS(language='KR', device=device),
|
15 |
}
|
16 |
speaker_ids = models['EN'].hps.data.spk2id
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
def synthesize(speaker, text, speed, language, progress=gr.Progress()):
|
18 |
bio = io.BytesIO()
|
19 |
models[language].tts_to_file(text, models[language].hps.data.spk2id[speaker], bio, speed=speed, pbar=progress.tqdm, format='wav')
|
20 |
return bio.getvalue()
|
21 |
-
def load_speakers(language):
|
22 |
-
|
|
|
|
|
|
|
|
|
23 |
with gr.Blocks() as demo:
|
24 |
-
gr.Markdown('# MeloTTS\n\
|
25 |
with gr.Group():
|
26 |
speaker = gr.Dropdown(speaker_ids.keys(), interactive=True, value='EN-Default', label='Speaker')
|
27 |
language = gr.Radio(['EN', 'ES', 'FR', 'ZH', 'JP', 'KR'], label='Language', value='EN')
|
28 |
-
language.input(load_speakers, inputs=language, outputs=speaker)
|
29 |
speed = gr.Slider(label='Speed', minimum=0.1, maximum=10.0, value=1.0, interactive=True, step=0.1)
|
30 |
-
text = gr.Textbox(label="Text to speak", value='
|
|
|
31 |
btn = gr.Button('Synthesize', variant='primary')
|
32 |
aud = gr.Audio(interactive=False)
|
33 |
btn.click(synthesize, inputs=[speaker, text, speed, language], outputs=[aud])
|
34 |
-
gr.Markdown('
|
35 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# WebUI by mrfakename <X @realmrfakename / HF @mrfakename>
|
2 |
+
# Demo also available on HF Spaces: https://huggingface.co/spaces/mrfakename/MeloTTS
|
3 |
import gradio as gr
|
4 |
import os, torch, io
|
5 |
+
# os.system('python -m unidic download')
|
6 |
+
print("Make sure you've downloaded unidic (python -m unidic download) for this WebUI to work.")
|
7 |
from melo.api import TTS
|
8 |
speed = 1.0
|
9 |
import tempfile
|
10 |
+
import click
|
11 |
+
device = 'auto'
|
12 |
models = {
|
13 |
'EN': TTS(language='EN', device=device),
|
14 |
'ES': TTS(language='ES', device=device),
|
|
|
18 |
'KR': TTS(language='KR', device=device),
|
19 |
}
|
20 |
speaker_ids = models['EN'].hps.data.spk2id
|
21 |
+
|
22 |
+
default_text_dict = {
|
23 |
+
'EN': 'The field of text-to-speech has seen rapid development recently.',
|
24 |
+
'ES': 'El campo de la conversión de texto a voz ha experimentado un rápido desarrollo recientemente.',
|
25 |
+
'FR': 'Le domaine de la synthèse vocale a connu un développement rapide récemment',
|
26 |
+
'ZH': 'text-to-speech 领域近年来发展迅速',
|
27 |
+
'JP': 'テキスト読み上げの分野は最近急速な発展を遂げています',
|
28 |
+
'KR': '최근 텍스트 음성 변환 분야가 급속도로 발전하고 있습니다.',
|
29 |
+
}
|
30 |
+
|
31 |
def synthesize(speaker, text, speed, language, progress=gr.Progress()):
|
32 |
bio = io.BytesIO()
|
33 |
models[language].tts_to_file(text, models[language].hps.data.spk2id[speaker], bio, speed=speed, pbar=progress.tqdm, format='wav')
|
34 |
return bio.getvalue()
|
35 |
+
def load_speakers(language, text):
|
36 |
+
if text in list(default_text_dict.values()):
|
37 |
+
newtext = default_text_dict[language]
|
38 |
+
else:
|
39 |
+
newtext = text
|
40 |
+
return gr.update(value=list(models[language].hps.data.spk2id.keys())[0], choices=list(models[language].hps.data.spk2id.keys())), newtext
|
41 |
with gr.Blocks() as demo:
|
42 |
+
gr.Markdown('# MeloTTS WebUI\n\nA WebUI for MeloTTS.')
|
43 |
with gr.Group():
|
44 |
speaker = gr.Dropdown(speaker_ids.keys(), interactive=True, value='EN-Default', label='Speaker')
|
45 |
language = gr.Radio(['EN', 'ES', 'FR', 'ZH', 'JP', 'KR'], label='Language', value='EN')
|
|
|
46 |
speed = gr.Slider(label='Speed', minimum=0.1, maximum=10.0, value=1.0, interactive=True, step=0.1)
|
47 |
+
text = gr.Textbox(label="Text to speak", value=default_text_dict['EN'])
|
48 |
+
language.input(load_speakers, inputs=[language, text], outputs=[speaker, text])
|
49 |
btn = gr.Button('Synthesize', variant='primary')
|
50 |
aud = gr.Audio(interactive=False)
|
51 |
btn.click(synthesize, inputs=[speaker, text, speed, language], outputs=[aud])
|
52 |
+
gr.Markdown('WebUI by [mrfakename](https://twitter.com/realmrfakename).')
|
53 |
+
@click.command()
|
54 |
+
@click.option('--share', '-s', is_flag=True, show_default=True, default=False, help="Expose a publicly-accessible shared Gradio link usable by anyone with the link. Only share the link with people you trust.")
|
55 |
+
@click.option('--host', '-h', default=None)
|
56 |
+
@click.option('--port', '-p', default=None)
|
57 |
+
def main(share, host, port):
|
58 |
+
demo.queue(api_open=False, default_concurrency_limit=10).launch(show_api=False, share=share, server_name=host, server_port=port)
|
59 |
+
|
60 |
+
if __name__ == "__main__":
|
61 |
+
main()
|