| from pprint import pprint | |
| from time import time | |
| from faster_whisper import WhisperModel | |
| model = WhisperModel("kotoba-tech/kotoba-whisper-v1.0-faster") | |
| test_audio = [ | |
| "kotoba-whisper-eval/audio/long_interview_1.wav", | |
| "kotoba-whisper-eval/audio/manzai1.wav", | |
| "kotoba-whisper-eval/audio/manzai2.wav", | |
| "kotoba-whisper-eval/audio/manzai3.wav" | |
| ] | |
| result = {} | |
| for test_audio_file in test_audio: | |
| start = time() | |
| segments, info = model.transcribe(test_audio_file, language="ja", chunk_length=15, condition_on_previous_text=False) | |
| for segment in segments: | |
| print("[%.2fs -> %.2fs] %s" % (segment.start, segment.end, segment.text)) | |
| elapsed = time() - start | |
| result[test_audio_file] = elapsed | |
| pprint(result) | |