Unose-Huen-Base (v0.5)
Unose-Huen-Base is a fine-tuned Whisper Base model (74M parameters) optimized for Nigerian English speech recognition, conversational prosody, and colloquial cadence.
Developed by Igunosa Labs as part of the Unose speech suite.
Model Details
- Architecture: Whisper Base (Encoder-Decoder Transformer, 6 layers, 74M params)
- Validation WER: 48.38% on the Nigerian Accented English benchmark
- Inference Targets: CPU / GPU / edge
- Available Formats: PyTorch safetensors, CTranslate2 (, , , )
Ecosystem Links
- Official GitHub Codebase: https://github.com/Igunosa-Labs/unose-huen
- Sister Model (Small 244M): igunosalabs/Unose-Huen-Small
- TTS Model: igunosalabs/Unose-Voice-v1.0
- Downloads last month
- -
Model tree for igunosalabs/Unose-Huen-Base
Base model
openai/whisper-base