mirror of https://github.com/k2-fsa/icefall.git synced 2025-12-09 14:05:33 +00:00

History

Fix setting joiner dim (#2027 )

Fixes incorrect computation of encoder_dim when encoder_dim is a comma-separated list of integers by ensuring numeric (not lexicographic) max is used.

Fixes #2018

- Replace int(max(params.encoder_dim.split(","))) (lexicographic max on strings) with max(_to_int_tuple(params.encoder_dim)) (numeric max).
- Apply the fix consistently across all affected training scripts.

2025-09-19 09:42:41 +08:00

local

fix str2bool (#1792 )

2024-10-31 09:54:12 +08:00

zipformer

Revert "Replace deprecated pytorch methods (#1814 )" (#1841 )

2024-12-18 16:49:57 +08:00

zipformer_prompt_asr

Fix setting joiner dim (#2027 )

2025-09-19 09:42:41 +08:00

prepare_prompt_asr.sh

PromptASR for contextualized ASR with controllable style (#1250 )

2023-10-11 14:56:41 +08:00

prepare.sh

Refactor data preparation for GigaSpeech recipe (#1986 )

2025-07-10 11:17:37 +08:00

README.md

Libriheavy recipe (zipformer) (#1261 )

2023-11-23 01:22:57 +08:00

RESULTS.md

Libriheavy recipe (zipformer) (#1261 )

2023-11-23 01:22:57 +08:00

shared

PromptASR for contextualized ASR with controllable style (#1250 )

2023-10-11 14:56:41 +08:00

README.md

Libriheavy: a 50,000 hours ASR corpus with punctuation casing and context

Libriheavy is a labeled version of Librilight. Please refer to our repository k2-fsa/libriheavy for more details. We also have a paper: Libriheavy: a 50,000 hours ASR corpus with punctuation casing and context, Preprint available on arxiv.

See RESULTS for the results for icefall recipes.