Synthetic Only Whisper Medium Synthetic Only Medium Lora Constwarmup Best7000 LoRA Adapter

Summary

This repository contains a Whisper checkpoint for Chichewa/Nyanja automatic speech recognition, fine-tuned from openai/whisper-medium.

• Experiment type: synthetic-only
• Base model: openai/whisper-medium
• Training condition: synthetic_only_medium_lora_constwarmup_best7000
• Release artifact: LoRA adapter checkpoint selected from the best training checkpoint

Intended use

This adapter is intended for research and evaluation on Chichewa/Nyanja ASR. It must be used together with the base Whisper model. It is not a production-ready speech system and should be validated carefully before downstream use.

How to use

This repository contains an adapter, not a fully merged standalone model. Load the base model first, then attach the adapter.

from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor
from peft import PeftModel

base_model_id = "openai/whisper-medium"
adapter_repo_id = "ai4good-labyrinth/synthetic-only-whisper-medium-lora-r32-no-language-v2"

processor = AutoProcessor.from_pretrained(base_model_id)
base_model = AutoModelForSpeechSeq2Seq.from_pretrained(base_model_id)
model = PeftModel.from_pretrained(base_model, adapter_repo_id)

The local evaluation script in this repository can also load the adapter directly because it reads adapter_config.json and automatically fetches the base model.

Training data

• Training source: Prepared dataset
• Evaluation source during training: FLEURS dev
• Train examples before duration filtering: 374612
• Train examples after duration filtering: 374576
• 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-medium
• Task: transcribe
• Language hint during training/evaluation: none, corresponding to --language auto in standalone evaluation
• LoRA: yes
• LoRA rank: 32
• LoRA alpha: 64
• LoRA dropout: 0.05
• LoRA target modules: k_proj,fc2,out_proj,v_proj,fc1,q_proj
• Extra trainable modules: embed_tokens,proj_out
• Mixed precision: fp16
• Gradient checkpointing: True
• Selected checkpoint step: 7000
• Selected checkpoint epoch: 0.15

Training-time dev selection

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

• Dev WER: 0.7020
• Dev CER: 0.2136
• Dev loss: 1.5151

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

Files in this repository

• Adapter 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.
• This repository contains an adapter only, so users must also comply with the upstream base model license and dataset licenses.
• 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

@misc{synthetic_only_whisper_medium_lora_r32_no_language_v2_2026,
  title        = {Synthetic Only Whisper Medium Synthetic Only Medium Lora Constwarmup Best7000 LoRA Adapter},
  author       = {AI4Good Labyrinth Team},
  year         = {2026},
  howpublished = {\url{https://huggingface.co/ai4good-labyrinth/synthetic-only-whisper-medium-lora-r32-no-language-v2}},
  note         = {Whisper LoRA adapter fine-tuning for Chichewa/Nyanja ASR}
}