Whisper Small – Nepali + English Code-Switching (Nepali Voice)

Fine-tuned on 98 high-quality clips (~60% Nepali, 25% mixed Nepali-English, 15% English) from speaker.

Perfect for transcribing natural Nepali conversations with English words mixed in.

Usage

from transformers import pipeline
asr = pipeline("automatic-speech-recognition", model="mandeep111/whisper-nepali-audio")
asr("your_nepali_audio.wav")
Downloads last month
6
Safetensors
Model size
0.2B params
Tensor type
F32
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support