Spaces:
Running
Running
File size: 840 Bytes
b0f692a cc8155a b0f692a cc8155a b0f692a | 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 | ---
title: Indigenous Language Classifier
emoji: π
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 6.6.0
app_file: app.py
pinned: false
license: mit
short_description: Tachiwin Indigenous Language Classifier (Audio)
---
# π Indigenous Language Classifier
Identify language family, superlanguage, and specific language code from audio files or recordings.
## π How to use
1. Upload an audio file (WAV or MP3) or use the microphone to record.
2. Click **Classify Language**.
3. View the results in the hierarchical classification dashboard.
## π Model Details
- **Architecture**: Whisper Encoder-Only Classifier
- **Scope**: 300+ indigenous languages
- **Region**: Mesoamerica and beyond
- **Metrics**:
- Family: ~73%
- Superlanguage: ~59%
- Code: ~52%
Created by [Tachiwin](https://huggingface.co/tachiwin).
|