gu-medium (CTranslate2)
A CTranslate2 conversion of vasista22/whisper-gujarati-medium for faster-whisper, as downloaded by LiveWhisper, a voice dictation app for Windows.
All credit for the model goes to its original authors. This repository only
changes the file format (float16 weights, plus tokenizer.json and
preprocessor_config.json that faster-whisper needs). The licence is the
original's: apache-2.0.
Clearly better than large-v3, but still misses about half the words. Whisper-medium fine-tuned by SPRING Lab, IIT Madras; 1.5 GB converted.
Measured in LiveWhisper
| metric | result |
|---|---|
| FLEURS word error | 49.8% (large-v3 67.7%) |
| delivered Gujlish | 48.5% (large-v3 63.3%) |
from faster_whisper import WhisperModel
model = WhisperModel("itsskofficial/livewhisper-gu-medium", device="cuda", compute_type="int8_float16")
- Downloads last month
- 13
Model tree for itsskofficial/livewhisper-gu-medium
Base model
vasista22/whisper-gujarati-medium