Experiment type: fleurs-synthetic
Base model: openai/whisper-small
Training condition: fleurs_synthetic_hours1p0
Release artifact: full fine-tuned checkpoint selected from the best training checkpoint

Intended use

This checkpoint is intended for research and evaluation on Chichewa/Nyanja ASR. It is not a production-ready speech system and should be validated carefully before downstream use.

How to use

This repository contains a full fine-tuned checkpoint. It can be loaded directly with Transformers.

python

from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor

model_id = "ai4good-labyrinth/fleurs-synthetic-hours1p0-whisper-small-no-language"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForSpeechSeq2Seq.from_pretrained(model_id)

Training data

Training source: FLEURS train + Prepared dataset
Evaluation source during training: FLEURS dev
Train examples before duration filtering: 10078
Train examples after duration filtering: 10037
Dev examples before duration filtering: 311
Dev examples after duration filtering: 305
Duration filter used during training: min_duration_seconds=0.0, max_duration_seconds=30.0

Training procedure

Fine-tuning script: experiments/whisper_finetune/finetune_whisper.py
Base model: openai/whisper-small
Task: transcribe
Language hint during training/evaluation: none, corresponding to --language auto in standalone evaluation
Mixed precision: no
Gradient checkpointing: False
Selected checkpoint step: 5200
Selected checkpoint epoch: 4.14

Training-time dev selection

The best checkpoint was selected using trainer-side dev evaluation on the duration-filtered FLEURS dev split.

Dev WER: 0.4854
Dev CER: 0.1444
Dev loss: 0.7309

These values come from the training pipeline and may differ slightly from standalone post-hoc evaluation because the decoding path is not perfectly identical.

Evaluation protocol

Standalone evaluation is recommended for the final release. Filtered and unfiltered results should be reported separately.

Filtered evaluation: min_duration_seconds=0, max_duration_seconds=30
Unfiltered evaluation: no duration constraint
Decoding task: transcribe
Language hint: auto

Evaluation summary

Dataset	Split	Setting	Num examples	WER	CER	Notes
FLEURS	dev	filtered	305	0.4995	0.1390	Filtered to 30 seconds
FLEURS	dev	unfiltered	TBD	TBD	TBD	Standalone eval pending
FLEURS	test	filtered	745	0.5028	0.1523	Filtered to 30 seconds
FLEURS	test	unfiltered	TBD	TBD	TBD	Standalone eval pending
Zambezi	dev	filtered	613	0.6468	0.2121	Filtered to 30 seconds
Zambezi	dev	unfiltered	TBD	TBD	TBD	Standalone eval pending
Zambezi	test	filtered	427	0.6348	0.1684	Filtered to 30 seconds
Zambezi	test	unfiltered	TBD	TBD	TBD	Standalone eval pending

Files in this repository

Model weights and config: repository root
Processor/tokenizer files: repository root
Evaluation JSON files: eval/...

Known limitations

Whisper does not provide an official Nyanja/Chichewa language token.
Users must also comply with the upstream dataset licenses and any upstream model license obligations.
Standalone evaluation and trainer-side evaluation can differ slightly even on the same split and duration filter.
Cross-dataset results should be interpreted carefully because transcription conventions may differ across corpora.

Citation

If you use this checkpoint, please cite:

the Whisper paper
the FLEURS dataset
this repository

bibtex

@misc{fleurs_synthetic_hours1p0_whisper_small_no_language_2026,
  title        = {Fleurs Synthetic Whisper Small Fleurs Synthetic Hours1P0 Checkpoint},
  author       = {AI4Good Labyrinth Team},
  year         = {2026},
  howpublished = {\url{https://huggingface.co/ai4good-labyrinth/fleurs-synthetic-hours1p0-whisper-small-no-language}},
  note         = {Whisper fine-tuning for Chichewa/Nyanja ASR}
}