whisper-large-v3-it-cv16-ct2

CTranslate2 conversion of fsicoli/whisper-large-v3-it-cv16 for use with faster-whisper.

Usage

With faster-whisper

from faster_whisper import WhisperModel

model = WhisperModel("fuzziness/whisper-large-v3-it-cv16-ct2", device="cuda", compute_type="float16")
segments, info = model.transcribe("audio.wav", language="it")

for segment in segments:
    print(f"[{segment.start:.2f}s -> {segment.end:.2f}s] {segment.text}")

With Speaches

STT_MODELS: fuzziness/whisper-large-v3-it-cv16-ct2

Conversion

Converted using:

ct2-transformers-converter --model fsicoli/whisper-large-v3-it-cv16 --output_dir . --quantization float16

Original Model

This is a CTranslate2 conversion of fsicoli/whisper-large-v3-it-cv16, a Whisper Large V3 model fine-tuned on Italian Common Voice 16 dataset.

All credit for the original training goes to fsicoli.