hassan-webm

Different outputs between file-based and mic-based streaming inference

1

Hi, I’m trying to simulate file streaming for a NeMo streaming ASR model. Specifically this one [nvidia/stt_en_fastconformer_hybrid_large_streaming_multi](https://huggingface.co/nvidia/stt_en_fastconformer_hybrid_large_streaming_multi) I noticed that: 1. While using [examples/asr/asr_cache_aware_streaming/speech_to_text_cache_aware_streaming_infer.py](https://github.com/NVIDIA-NeMo/NeMo/blob/main/examples/asr/asr_cache_aware_streaming/speech_to_text_cache_aware_streaming_infer.py), which uses `CacheAwareStreamingAudioBuffer` and preprocesses the...