KasuleTrevor/cdli-qwen3-asr-lg-atypical-stage3-1p7b-typical-base-spec
07
Qwen3-ASR Luganda Atypical Stage 3 1.7B SpecAugment
This repository contains a Qwen3-ASR checkpoint fine-tuned for Luganda atypical speech from the CDLI Luganda non-standard speech dataset, using the SpecAugment-enabled training notebook.
Training Setup
- Base model:
KasuleTrevor/cdli-qwen3-asr-lg-typical-1p7b-base-finetune - Dataset:
cdli/ugandan_luganda_nonstandard_speech_v1.0 - Language: Luganda (
lg) - Training source notebook:
qwen3-asr-luganda-atypical-cleaned-finetune-spec.ipynb - Learning rate:
5e-5 - Scheduler:
constant_with_warmup - Epochs:
3 - Batch size:
2 - Gradient accumulation:
4 - Save steps:
250 - SpecAugment: enabled
Checkpoint Selection
Notebook outputs show that validation checkpoint comparison selected checkpoint-500 by normalized corpus WER.
Test Results
Per-Speaker Breakdown
These per-speaker aggregates come from the notebook-displayed grouped analysis for checkpoint-500.
<!-- markdownlint-disable MD013 -->
<!-- markdownlint-enable MD013 -->
Evaluation Notes
The model card is reconstructed from executed notebook outputs and extracted local summaries. Full per-utterance prediction files were referenced in notebook outputs as remote /workspace/... artifact paths, but were not embedded in the notebook JSON available locally.
