o i @s8dZddlmZmZmZddlZGdddejjZdS)z%MultiBlocks for encoder architecture.)DictListOptionalNc seZdZdZejjdfdeejjde dejjde e ddf fdd Z d e d ej ddfd d Z ddejdejdejde ejdejf ddZ  ddejdejdejd e de dejf ddZZS) MultiBlockszMultiBlocks definition. Args: block_list: Individual blocks of the encoder architecture. output_size: Architecture output size. norm_class: Normalization module class. norm_args: Normalization module arguments. N block_list output_size norm_class norm_argsreturncs8ttj||_||fi||_t||_dS)zConstruct a MultiBlocks object.N) super__init__torchnn ModuleListblocks norm_blockslen num_blocks)selfrrrr  __class__g/home/ubuntu/.local/lib/python3.10/site-packages/espnet2/asr_transducer/encoder/modules/multi_blocks.pyr s zMultiBlocks.__init__ left_contextdevicecCs&t|jD] }|j|||qdS)zInitialize/Reset encoder streaming cache. Args: left_context: Number of left frames during chunk-by-chunk inference. device: Device to use for cache tensor. N)rangerrreset_streaming_cache)rrridxrrrr"sz!MultiBlocks.reset_streaming_cachexpos_encmask chunk_maskcCs8t|jD]\}}|||||d\}}}q||}|S)aUForward each block of the encoder architecture. Args: x: MultiBlocks input sequences. (B, T, D_block_1) pos_enc: Positional embedding sequences. mask: Source mask. (B, T) chunk_mask: Chunk mask. (T_2, T_2) Returns: x: Output sequences. (B, T, D_block_N) )r!) enumeraterr)rrrr r! block_indexblockrrrforward-s zMultiBlocks.forwardr right_contextcCs:t|jD]\}}|j|||||d\}}q||}|S)aForward each block of the encoder architecture. Args: x: MultiBlocks input sequences. (B, T, D_block_1) pos_enc: Positional embedding sequences. (B, 2 * (T - 1), D_att) mask: Source mask. (B, T_2) left_context: Number of frames in left context. right_context: Number of frames in right context. Returns: x: MultiBlocks output sequences. (B, T, D_block_N) )rr&)r"r chunk_forwardr)rrrr rr& block_idxr$rrrr'Gs zMultiBlocks.chunk_forward)N)rr)__name__ __module__ __qualname____doc__r r LayerNormrModuleintrrr rrTensorr%r' __classcell__rrrrrsV  r) r,typingrrrr rr.rrrrrs