Trelis/whisper-turbo-atc-25h

Whisper Turbo Atc 25H

Fine-tuned Whisper model based on openai/whisper-large-v3-turbo.

Training Results

Metric	Base Model	Fine-tuned
WER	60.73%	18.92%

Improvement: 41.81% WER reduction (lower is better)

Training Details

• Base Model: openai/whisper-large-v3-turbo
• Training Dataset: Trelis/atc-train-25h
• Train Loss: 0.2888
• Training Time: 10.9 minutes

Inference

from transformers import pipeline

asr = pipeline("automatic-speech-recognition", model="Trelis/whisper-turbo-atc-25h")
result = asr("path/to/audio.wav")
print(result["text"])

Training Logs

Full training logs are available in training_log.txt.

---

Fine-tuned using Trelis Studio