File size: 3,770 Bytes
b59e10d 6616e45 b59e10d |
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 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 |
import gradio as gr
import tempfile
from TTS.utils.synthesizer import Synthesizer
from huggingface_hub import hf_hub_download
REPO_ID = "mbarnig/lb-de-fr-en-pt-coqui-vits-tts"
my_title = "🇩🇪 🇫🇷 🇬🇧 🇵🇹 Mir schwätzen och Lëtzebuergesch ! 🇱🇺"
my_description = "First multilingual-multispeaker Text-to-Speech (TTS) synthesizer speaking Luxembourgish. This model is based on [YourTTS](https://github.com/Edresson/YourTTS), thanks to 🐸 [Coqui.ai](https://coqui.ai/)."
lb_text = "An der Zäit hunn sech den Nordwand an d'Sonn gestridden, wie vun hinnen zwee wuel méi staark wier, wéi e Wanderer, deen an ee waarme Mantel agepak war, iwwert de Wee koum."
de_text = "Einst stritten sich Nordwind und Sonne, wer von ihnen beiden wohl der Stärkere wäre, als ein Wanderer, der in einen warmen Mantel gehüllt war, des Weges daherkam."
fr_text = "La bise et le soleil se disputaient, chacun assurant qu'il était le plus fort, quand ils ont vu un voyageur qui s'avançait, enveloppé dans son manteau."
en_text = "The North Wind and the Sun were disputing which was the stronger, when a traveler came along wrapped in a warm cloak."
pt_text = "O vento norte e o Sol discutiam quem era o mais forte, quando surgiu um viajante envolvido numa capa."
TTS_VOICES = [
"Judith",
"Luc",
"Kerstin",
"Ed",
"Linda"
]
TTS_LANGUAGES = [
"x-lb",
"x-de",
"fr-fr",
"en",
"pt-br"
]
my_examples = [
[lb_text,"x-lb","Luc"],
[de_text,"x-de","Judith"],
[fr_text,"fr-fr","Kerstin"],
[en_text,"en","Ed"],
[pt_text,"pt-br","Linda"],
]
my_article = "<h3>User guide</h3><p>1. Press the Submit button to generate a speech file with the default values. 2. Change the default values by clicking an example row. 3. Select a language and a voice and enter your own text. Have fun!</p><p>Go to <a href='https://www.web3.lu/category/audio-technologies/'>Internet with a Brain</a> to read some technical infos.</p>"
my_inputs = [
gr.inputs.Textbox(lines=5, label="Input Text", default=lb_text),
gr.inputs.Radio(label="Select a language", choices = TTS_LANGUAGES, default = "x-lb"),
gr.inputs.Radio(label="Select a voice", choices = TTS_VOICES, default = "Judith"),
]
my_outputs = gr.outputs.Audio(type="file", label="Output Audio")
def tts(text: str, speaker_idx: str, language_idx: str):
best_model_path = hf_hub_download(repo_id=REPO_ID, filename="best_model.pth")
print(best_model_path)
config_path = hf_hub_download(repo_id=REPO_ID, filename="config.json")
print(config_path)
speakers_path = hf_hub_download(repo_id=REPO_ID, filename="speakers.pth")
print(speakers_path)
languages_path = hf_hub_download(repo_id=REPO_ID, filename="language_ids.json")
print(languages_path)
speaker_encoder_model_path = hf_hub_download(repo_id=REPO_ID, filename="model_se.pth")
print(speaker_encoder_model_path)
speaker_encoder_config_path = hf_hub_download(repo_id=REPO_ID, filename="config_se.json")
print(speaker_encoder_config_path)
# init synthesizer
synthesizer = Synthesizer(
best_model_path,
config_path,
speakers_path,
languages_path,
None,
None,
speaker_encoder_model_path,
speaker_encoder_config_path,
False
)
# create audio file
wavs = synthesizer.tts(text, speaker_idx, language_idx)
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 = my_article,
examples = my_examples,
allow_flagging=False
)
iface.launch()
|