icefall
icefall copied to clipboard
pad_length in streaming decode
Hi,
I'm confused about the pad_length in set_features. Since the frames that are less than chunk_size*2+7+2*3 have already been padded in streaming_decode.py, why are we still adding 7+2*3 when initially receiving the feature here? Is the 7+2*3 necessary in set_features?
https://github.com/k2-fsa/icefall/blob/cea0dbe7b1cd4d5b7512c7974e53034ef456dd70/egs/librispeech/ASR/zipformer/decode_stream.py#L108-L121
https://github.com/k2-fsa/icefall/blob/cea0dbe7b1cd4d5b7512c7974e53034ef456dd70/egs/librispeech/ASR/zipformer/streaming_decode.py#L472-L484
Thanks!