KasuleTrevor/cdli-qwen3-asr-lg-typical-finetune-yv

CDLI Qwen3-ASR Luganda Typical Speech Fine-tune

This repo contains checkpoint: checkpoint-3165.

Training Setup

• Base model: Qwen/Qwen3-ASR-1.7B
• Datasets: KasuleTrevor/lg_100hrs + dmusingu/yogera-dataset (Luganda split)
• Task: Automatic Speech Recognition
• Language: Luganda

Notes

• This upload excludes training-state files to keep the repo clean.
• Checkpoint folder includes tokenizer and processor files needed for inference.
• Uploaded: 2026-04-11T02:20:58.125883+00:00

Checkpoint Metrics (Normalized)

• WER: 0.316801
• CER: 0.077357
• Evaluated samples: 2828