icefall

mirror of https://github.com/k2-fsa/icefall.git synced 2025-12-11 06:55:27 +00:00

History

Enabling char_level and compute_CER for aishell recipe (#1554 )

* init fix

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

2024-03-18 11:57:47 +08:00

asr_datamodule.py

…

beam_search.py

…

decode_stream.py

…

decode.py

Enabling char_level and compute_CER for aishell recipe (#1554 )

2024-03-18 11:57:47 +08:00

decoder.py

…

do_not_use_it_directly.py

…

encoder_interface.py

…

export-for-ncnn-zh.py

…

export-for-ncnn.py

…

export-onnx-zh.py

…

export-onnx.py

…

export.py

…

jit_pretrained.py

…

jit_trace_export.py

…

jit_trace_pretrained.py

…

joiner.py

…

model.py

…

ncnn_custom_layer.py

…

onnx_check.py

…

onnx_model_wrapper.py

…

onnx_pretrained.py

…

optim.py

…

pretrained.py

…

README.md

…

scaling_converter.py

…

scaling.py

…

streaming_beam_search.py

…

streaming_decode.py

…

streaming-ncnn-decode.py

…

test_model.py

…

train.py

…

zipformer_for_ncnn_export_only.py

…

zipformer.py

…

README.md

This recipe implements Streaming Zipformer-Transducer model.

See https://k2-fsa.github.io/icefall/recipes/Streaming-ASR/librispeech/zipformer_transducer.html for detailed tutorials.

./emformer.py and ./train.py are basically the same as ./emformer2.py and ./do_not_use_it_directly.py. The only purpose of ./emformer2.py and ./do_not_use_it_directly.py is for exporting to sherpa-ncnn.