|
|
---
|
|
|
title: veureu-asr
|
|
|
emoji: 🗣️
|
|
|
colorFrom: pink
|
|
|
colorTo: pink
|
|
|
sdk: gradio
|
|
|
sdk_version: "4.44.1"
|
|
|
app_file: app.py
|
|
|
pinned: false
|
|
|
---
|
|
|
|
|
|
# 🗣️ veureu-asr (Aina faster-whisper · Català · ZeroGPU)
|
|
|
|
|
|
Reconocimiento de voz en catalán basado en **faster-whisper** (CTranslate2) con el modelo de **projecte-aina**.
|
|
|
|
|
|
## Endpoints (Gradio)
|
|
|
- **`/api/predict`** — entrada: `[ <audio_file>, "ca", true, true ]` → salida:
|
|
|
```json
|
|
|
{
|
|
|
"text": "…",
|
|
|
"segments": [{"start": 0.1, "end": 1.9, "text": "…"}],
|
|
|
"language": "ca",
|
|
|
"info": {"duration": 12.3, "device": "cuda", "compute_type": "float16"}
|
|
|
}
|
|
|
|
