Downloads · 30 days
22
10% of all-time downloads
Rekody/whisper-large-v3-turbo-kalenjin
whisper-large-v3-turbo-kalenjin is a automatic speech recognition model from Rekody. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as mit.
Kalenjin automatic speech recognition. This is openai/whisper-large-v3-turbo fine-tuned on Kalenjin speech (Kipsigis and Nandi dialects) with LoRA, then merged into the base weights. It loads like any regular Whisper…
Downloads · 30 days
22
10% of all-time downloads
All-time downloads
220
Public
Parameters
809M
1.6 GB on disk
Likes
2
Public
Click a slice to open those files.
.safetensors1.6 GB · 100%
From the Hugging Face model README
Kalenjin automatic speech recognition. This is openai/whisper-large-v3-turbo fine-tuned on Kalenjin speech (Kipsigis and Nandi dialects) with LoRA, then merged into the base weights. It loads like any regular Whisper checkpoint: no PEFT required.
This is v1 of this repository.
Scored on a frozen 198-clip held-out set with the canonical decoding recipe: 30 s chunks, stride (5, 5), beam size 5, temperature fallback.
| Model | CER | CER (orthography-normalized) |
|---|---|---|
| This model (v1) | 0.1959 | 0.1629 |
| Previous best Kalenjin Whisper fine-tune | 0.2170 | 0.1848 |
Both rows use the same evaluation harness. The improvement is significant under a paired bootstrap test (p = 0.005).
Trained on the scripted split of AfriVoices-KE (the Anv-ke/Kalenjin dataset, CC-BY-4.0; Wanzare et al., 2026), with online speed, noise, and gain augmentation plus SpecAugment. Fine-tuning used LoRA (rank 16, learning rate 2e-4); the adapter was merged into the base model for this release.
import torch
from transformers import pipeline
asr = pipeline(
"automatic-speech-recognition",
model="Rekody/whisper-large-v3-turbo-kalenjin",
torch_dtype=torch.float16,
device="cuda", # or "mps" on Apple silicon, or "cpu"
)
result = asr(
"audio.wav",
chunk_length_s=30,
stride_length_s=(5, 5),
generate_kwargs={"language": "sw", "task": "transcribe", "num_beams": 5},
)
print(result["text"])
language="sw"?Whisper's tokenizer has no Kalenjin language token. This model was trained with the Swahili token (sw) standing in for Kalenjin, so pass language="sw" at inference. The output is Kalenjin text.
MIT, matching the base model. The training dataset is separately licensed (CC-BY-4.0) under its own access and consent terms.
If you use this model, please cite:
Kipkemboi, T. (2026). Open Kalenjin Automatic Speech Recognition: Adapting Parakeet-TDT to a Low-Resource Nilotic Language. Zenodo. https://doi.org/10.5281/zenodo.21764656
For the dataset, cite Wanzare et al. (2026), AfriVoices-KE.
Training and evaluation code: https://github.com/rekody/kalenjin-asr