CoolFace
Modelpublic

LocalAI-io/whisper-tiny-it-multi-yodas-asr_only-nocv-ct2-int8

sourceHugging Facemitupdated 4mo agoView on Hugging Face
0likes8downloads
Model Card

whisper-tiny-it-multi-yodas-asr_only-nocv-ct2-int8

CTranslate2 INT8 quantized version of LocalAI-io/whisper-tiny-it-multi-yodas-asr_only-nocv for fast CPU inference.

Author: Ettore Di Giacinto

Brought to you by the LocalAI team. This model can be used directly with LocalAI.

Training

  • —Base model: openai/whisper-tiny, fine-tuned on MLS + VoxPopuli + YODAS-Granary (asr_only) Italian
  • —Quantization: INT8 via CTranslate2

Usage

faster-whisper

python
from faster_whisper import WhisperModel

model = WhisperModel("LocalAI-io/whisper-tiny-it-multi-yodas-asr_only-nocv-ct2-int8", device="cpu", compute_type="int8")
segments, info = model.transcribe("audio.mp3", language="it")
for segment in segments:
    print(f"[{segment.start:.1f}s - {segment.end:.1f}s] {segment.text}")

WhisperX

python
import whisperx

model = whisperx.load_model("LocalAI-io/whisper-tiny-it-multi-yodas-asr_only-nocv-ct2-int8", device="cpu", compute_type="int8")
result = model.transcribe("audio.mp3", language="it")

Links