kn-medium (CTranslate2)

A CTranslate2 conversion of vasista22/whisper-kannada-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.

Whisper-medium fine-tuned on Kannada by SPRING Lab, IIT Madras. 1.5 GB once converted; converting peaks near 7 GB of memory.

Measured in LiveWhisper

metric result
FLEURS word error 32.3% (large-v3 67.4%)
delivered Kanglish 30.2% (large-v3 59.8%)
from faster_whisper import WhisperModel
model = WhisperModel("itsskofficial/livewhisper-kn-medium", device="cuda", compute_type="int8_float16")
Downloads last month
12
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for itsskofficial/livewhisper-kn-medium

Finetuned
(3)
this model