File size: 3,467 Bytes
3f4dd11 49152bc 3f4dd11 |
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 |
import gradio as gr
from transformers import VitsModel, AutoTokenizer
import torch
import logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
languages = ["bambara", "boomu", "dogon", "pular", "songhoy", "tamasheq"]
models = {}
tokenizers = {}
examples = {
"bambara": "An filɛ ni ye yɔrɔ minna ni an ye an sigi ka a layɛ yala an bɛ ka baara min kɛ ɛsike a kɛlen don ka Ɲɛ wa ?",
"boomu": "Vunurobe wozomɛ pɛɛ, Poli we zo woro han Deeɓenu wara li Deeɓenu faralo zuun. Lo we baba a lo wara yi see ɓa Zuwifera ma ɓa Gɛrɛkela wa.",
"dogon": "Pɔɔlɔ, kubɔ lugo joo le, bana dɛin dɛin le, inɛw Ama titiyaanw le digɛu, Ama, emɛ babe bɛrɛ sɔɔ sɔi.",
"pular": "Miɗo ndaarde saabe Laamɗo e saabe Iisaa Almasiihu caroyoowo wuurɓe e maayɓe oo, miɗo ndaardire saabe gartol makko ka num e Laamu makko",
"songhoy": "Haya ka se beenediyo kokoyteraydi go hima nda huukoy foo ka fatta ja subaahi ka taasi goykoyyo ngu rezẽ faridi se",
"tamasheq": "Toḍă tăfukt ɣas, issăɣră-dd măssi-s n-ašĕkrĕš ănaẓraf-net, inn'-as: 'Ǝɣĕr-dd inaxdimăn, tĕẓlĕd-asăn, sănt s-wi dd-ĕšrăynen har tĕkkĕd wi dd-ăzzarnen."
}
try:
for lang in languages:
logger.info(f"Loading model and tokenizer for {lang}...")
models[lang] = VitsModel.from_pretrained("sudoping01/malian-tts", subfolder=f"models/{lang}")
tokenizers[lang] = AutoTokenizer.from_pretrained("sudoping01/malian-tts", subfolder=f"models/{lang}")
logger.info(f"Successfully loaded {lang}")
except Exception as e:
logger.error(f"Failed to load models: {str(e)}")
raise Exception(f"Model loading failed: {str(e)}")
def generate_audio(language, text):
if not text.strip():
return None, "Please enter some text to synthesize."
try:
model = models[language]
tokenizer = tokenizers[language]
inputs = tokenizer(text, return_tensors="pt")
with torch.no_grad():
output = model(**inputs).waveform
waveform = output.squeeze().cpu().numpy()
sample_rate = model.config.sampling_rate
return (sample_rate, waveform), None
except Exception as e:
logger.error(f"Error during inference for {language}: {str(e)}")
return None, f"Error generating audio: {str(e)}"
def load_example(language):
return examples.get(language, "No example available")
with gr.Blocks(title="Malian Languages TTS") as demo:
gr.Markdown("# Malian Languages Text-to-Speech")
gr.Markdown("Select a language, enter text or load an example, and listen to the synthesized speech!")
with gr.Row():
language = gr.Dropdown(choices=languages, label="Language", value="bambara")
with gr.Column():
text = gr.Textbox(label="Input Text", lines=5, placeholder="Type your text here...")
example_btn = gr.Button("Load Example")
generate_btn = gr.Button("Generate Audio", variant="primary")
audio_output = gr.Audio(label="Generated Audio", type="numpy")
error_msg = gr.Textbox(label="Status", visible=False)
# Connect buttons to functions
generate_btn.click(
fn=generate_audio,
inputs=[language, text],
outputs=[audio_output, error_msg]
)
example_btn.click(
fn=load_example,
inputs=language,
outputs=text
)
demo.launch() |