Zengwei Yao ece728d895

Apply delay penalty on k2 ctc loss (#669) ...

* add init files

* fix bug, apply delay penalty

* fix decoding code and getting timestamps

* add option applying delay penalty on ctc log-prob

* fix bug of streaming decoding

* minor change for bpe-based case

* add test_model.py

* add README.md

* add CI

2022-11-28 22:34:02 +08:00

..

compute-fbank-librispeech-test-clean-and-test-other.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

download-gigaspeech-dev-test-dataset.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

download-librispeech-test-clean-and-test-other-dataset.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

install-kaldifeat.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

prepare-librispeech-test-clean-and-test-other-manifests.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-aishell-pruned-transducer-stateless3-2022-06-20.sh

Support exporting LSTM with projection to ONNX (#621)

2022-10-18 11:25:31 +08:00

run-gigaspeech-pruned-transducer-stateless2-2022-05-12.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-librispeech-conformer-ctc3-2022-11-28.sh

Apply delay penalty on k2 ctc loss (#669)

2022-11-28 22:34:02 +08:00

run-librispeech-lstm-transducer-stateless2-2022-09-03.yml

Fix shallow fusion and add CI tests for it (#676)

2022-11-13 11:51:00 +08:00

run-librispeech-pruned-transducer-stateless2-2022-04-29.sh

Add fast_beam_search_LG (#622)

2022-11-03 16:29:30 +08:00

run-librispeech-pruned-transducer-stateless3-2022-04-29.sh

Add fast_beam_search_LG (#622)

2022-11-03 16:29:30 +08:00

run-librispeech-pruned-transducer-stateless3-2022-05-13.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-librispeech-pruned-transducer-stateless5-2022-05-13.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-librispeech-pruned-transducer-stateless7-2022-11-11.sh

Add zipformer from Dan using multi-dataset setup (#675)

2022-11-15 16:56:05 +08:00

run-librispeech-pruned-transducer-stateless8-2022-11-14.sh

Add zipformer from Dan using multi-dataset setup (#675)

2022-11-15 16:56:05 +08:00

run-librispeech-pruned-transducer-stateless-2022-03-12.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-librispeech-streaming-pruned-transducer-stateless2-2022-06-26.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-librispeech-transducer-stateless2-2022-04-19.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-pre-trained-conformer-ctc.sh

Modify ActivationBalancer for speed (#612)

2022-10-13 15:14:28 +08:00

run-pre-trained-transducer-stateless-librispeech-100h.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-pre-trained-transducer-stateless-librispeech-960h.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-pre-trained-transducer-stateless-modified-2-aishell.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-pre-trained-transducer-stateless-modified-aishell.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-pre-trained-transducer-stateless.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-pre-trained-transducer.sh

Remove all-in-one for onnx export (#614)

2022-10-12 10:34:06 +08:00

run-wenetspeech-pruned-transducer-stateless2.sh

Support exporting to ONNX for the wenetspeech recipe (#615)

2022-10-13 15:17:20 +08:00