| import sys | |
| from speechbrain.inference.ASR import StreamingASR | |
| from speechbrain.utils.dynamic_chunk_training import DynChunkTrainConfig | |
| asr_model = StreamingASR.from_hparams(source='speechbrain-stt', savedir='speechbrain-stt') | |
| def transcribe( audio_file, chunk_size=24, chunk_stride=4, use_torchaudio_streaming=False): | |
| result = asr_model.transcribe_file( | |
| audio_file, | |
| DynChunkTrainConfig(chunk_size, chunk_stride), | |
| use_torchaudio_streaming=use_torchaudio_streaming, | |
| ) | |
| return result | |
| audio_file = sys.argv[1] | |
| transcription = transcribe(audio_file) | |
| print(f"Transcription: {transcription}") | |