File size: 1,712 Bytes
be1bb68 |
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 |
from TTS.utils.synthesizer import Synthesizer
from huggingface_hub import hf_hub_download
import gradio as gr
import tempfile
REPO_ID = "jhlfrfufyfn/bel-tts"
my_title = "Беларускі тэкст-у-маўленне"
my_description = "Беларускамоўны мадэль для агучвання тэкста. "
be_text = "Гепарды жывуць у адкрытых і прасторных месцах, дзе ёсць шмат здабычы."
my_inputs = [
gr.inputs.Textbox(lines=5, label="Input Text", default=be_text),
]
my_outputs = gr.outputs.Audio(type="file", label="Output Audio")
def tts(text: str):
best_model_path = hf_hub_download(repo_id=REPO_ID, filename="model.pth")
config_path = hf_hub_download(repo_id=REPO_ID, filename="config.json")
vocoder_path = hf_hub_download(repo_id=REPO_ID, filename="vocoder.pth")
scale_stats_path = hf_hub_download(repo_id=REPO_ID, filename="scale_stats.npy")
vocoder_config_path = hf_hub_download(repo_id=REPO_ID, filename="vocoder_config.json")
# init synthesizer
synthesizer = Synthesizer(
best_model_path,
config_path,
None,
None,
vocoder_path,
vocoder_config_path,
None,
None,
False
)
# create audio file
wavs = synthesizer.tts(text)
with tempfile.NamedTemporaryFile(suffix = ".wav", delete = False) as fp:
synthesizer.save_wav(wavs, fp)
return fp.name
iface = gr.Interface(
fn=tts,
inputs=my_inputs,
outputs=my_outputs,
title=my_title,
description = my_description,
article = "",
examples = "",
allow_flagging=False
)
iface.launch()
|