Downloads · 30 days
3
14% of all-time downloads
whisper-small-hindi-lora
manishehehe/whisper-small-hindi-lora
whisper-small-hindi-lora is a automatic speech recognition model from manishehehe. Use it when you need speech turned into text. It is set up for peft. The card lists the license as mit.
LoRA adapter for openai/whisper-small, fine-tuned for Hindi ASR on 500 FLEURS (hiin) utterances. ~35 minutes on a free Colab T4.
All-time downloads
22
Public
Repo size
14.2 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors14.2 MB · 78%
At a glance
- Task
- Automatic Speech Recognition
- Library
- peft
- License
- mit
- Access
- Public
- Created
- Aug 23, 2026
- Updated
- Aug 23, 2026
- SHA
- b6574502
Base models
Datasets
Model card
From the Hugging Face model README
whisper-small-hindi-lora
LoRA adapter for openai/whisper-small, fine-tuned for Hindi ASR on 500 FLEURS (hi_in) utterances. ~35 minutes on a free Colab T4.
Results (IndicTTS eval set, out-of-domain)
| Voice | Baseline WER | LoRA WER | Relative Δ | Sarvam Saaras v3 |
|---|---|---|---|---|
| Hindi Female (n=30) | 0.515 | 0.362 | −29.8% | 0.237 |
Closes 55% of the gap to Sarvam Saaras v3 (0.278 → 0.125 WER points).
Usage
from transformers import WhisperForConditionalGeneration, WhisperProcessor
from peft import PeftModel
processor = WhisperProcessor.from_pretrained("openai/whisper-small", language="Hindi", task="transcribe")
model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-small")
model = PeftModel.from_pretrained(model, "manishehehe/whisper-small-hindi-lora")
model.eval()
ids = model.generate(input_features=input_features, language="hi", task="transcribe", max_new_tokens=225)
print(processor.tokenizer.batch_decode(ids, skip_special_tokens=True)[0])
Training
LoRA on q_proj/v_proj (r=32, α=64, dropout=0.05) — ~1.8M trainable params (0.7% of base). 500 steps, effective batch 32, LR 1e-3, fp16, single T4.
Code: https://github.com/manishehehe/whisper-indic-lora
Limitations
Small eval set (n=30); WER reductions are descriptive (no per-utterance significance test). FLEURS read speech differs in domain from IndicTTS studio recordings.
- Task
- Automatic Speech Recognition
- Library
- peft
- License
- mit
- Languages
- hi
- Created
- Aug 23, 2026
- Updated
- Aug 23, 2026