Commit Graph

  • fba5e67d5e
    Fix CI tests. (#1974) Fangjun Kuang 2025-07-01 13:47:55 +08:00
  • 127b4985d3 small fixes k2-fsa 2025-07-01 13:41:56 +08:00
  • b4e9edbed1 minor fixes k2-fsa 2025-07-01 13:31:03 +08:00
  • 82af46284f Merge branch 'fix-ci-2' into fix-ci k2-fsa 2025-07-01 11:41:26 +08:00
  • 633eec5445 small fixes k2-fsa 2025-07-01 11:14:48 +08:00
  • a91d890552 fix grad scaler k2-fsa 2025-07-01 00:05:08 +08:00
  • a1277c9ae9 fix grad scaler k2-fsa 2025-07-01 00:05:08 +08:00
  • f186e1d427 Fix weights_only=False k2-fsa 2025-06-30 22:07:36 +08:00
  • a53c323750 Fix CI warnings k2-fsa 2025-06-30 21:46:18 +08:00
  • ffe2f16b1d Fix librispeech CI test errors k2-fsa 2025-06-30 20:36:21 +08:00
  • fe36fcc25c Refactor CI k2-fsa 2025-06-30 19:04:02 +08:00
  • 71377d21cd
    Export streaming zipformer models with whisper feature to onnx (#1973) Fangjun Kuang 2025-06-30 19:01:15 +08:00
  • a318ac20c3 export fp16 onnx models Fangjun Kuang 2025-06-30 11:34:50 +08:00
  • 9d4b0dfcd4 Export multi_zh-hans models to onnx Fangjun Kuang 2025-06-30 10:58:18 +08:00
  • f8354ee64d
    Merge 36808b89406d97ee5ab68c43136a509eb0d193fc into abd9437e6d5419a497707748eb935e50976c3b7b Fangjun Kuang 2025-06-27 11:34:28 +00:00
  • 0ec57e77a2
    Merge e03a2d1d9048e0bdb40e28dfcc0d2fa72f6e8614 into abd9437e6d5419a497707748eb935e50976c3b7b Daniel Doña 2025-06-27 11:34:25 +00:00
  • 65275d5e48
    Merge 0eccb2b62cc7ed8f7066bcc2090984ff87535006 into abd9437e6d5419a497707748eb935e50976c3b7b Daniel Povey 2025-06-27 11:34:23 +00:00
  • d828e4b281
    Merge 3954dfeaba6809f25fe16324cce899f55e02c2e2 into abd9437e6d5419a497707748eb935e50976c3b7b Wen Ding 2025-06-27 11:34:02 +00:00
  • 57214969b1
    Merge 8809c7a99115277a2d4a77c69714ada661ba6653 into abd9437e6d5419a497707748eb935e50976c3b7b kobenaxie 2025-06-27 11:33:58 +00:00
  • d1aa1b06f5
    Merge 78ddda42967d018fe5f0f7746bef4eca630a1253 into abd9437e6d5419a497707748eb935e50976c3b7b Amir Hussein 2025-06-27 11:33:53 +00:00
  • c32cef02a5
    Merge 5beef022852e1dea957530b7f27d467eb7373415 into abd9437e6d5419a497707748eb935e50976c3b7b Zengwei Yao 2025-06-27 11:33:30 +00:00
  • d07ad97691
    Merge 7cdc0da3391e5c3cdd8b7ee9a9db52a7a1d6e641 into abd9437e6d5419a497707748eb935e50976c3b7b Peter Ross 2025-06-27 11:33:23 +00:00
  • 2c7dcd65f2
    Merge 4858e2b0367a7ebaf20641743d91a745777aca63 into abd9437e6d5419a497707748eb935e50976c3b7b Yifan Yang 2025-06-27 11:33:11 +00:00
  • b5ae175e8e
    Merge 0fb43289f477aa1a9f1d88215684f7808c1c0fd8 into abd9437e6d5419a497707748eb935e50976c3b7b Liyong.Guo 2025-06-27 11:33:00 +00:00
  • a84bd6ae1f
    Merge 1aa6fc0122981d3a248b08edd2bf4d71f5e27bd0 into abd9437e6d5419a497707748eb935e50976c3b7b Zengwei Yao 2025-06-27 11:33:00 +00:00
  • 0290083cfb
    Merge cc74ba574e341081fa60cfe5f3cdc1fc905a18df into abd9437e6d5419a497707748eb935e50976c3b7b Zengwei Yao 2025-06-27 11:32:55 +00:00
  • 261463cf27
    Merge 723320e0159d65856d3c86ae4f75b9d44034fc3d into abd9437e6d5419a497707748eb935e50976c3b7b Zengwei Yao 2025-06-27 11:32:43 +00:00
  • d5c3ac833c
    Merge 6e0133902f633dad5c9d28e040da660ed2c184e3 into abd9437e6d5419a497707748eb935e50976c3b7b Xiaoyu Yang 2025-06-27 11:32:43 +00:00
  • 930148c363
    Merge 903ef3b1612b1bec1a2ab682c5849f76cec9e6af into abd9437e6d5419a497707748eb935e50976c3b7b Fangjun Kuang 2025-06-27 11:32:28 +00:00
  • a0c0d2578f
    Merge fb897bdd77759f17368ed9d7a359fe7bc031f921 into abd9437e6d5419a497707748eb935e50976c3b7b Zengwei Yao 2025-06-27 11:32:18 +00:00
  • 7b225b660e
    Merge fb90ada9e86084926f228eaeb8e41287c50fc1cf into abd9437e6d5419a497707748eb935e50976c3b7b rickychanhoyin 2025-06-27 11:32:12 +00:00
  • 816614b50e
    Merge 890cd1ab7529a5d284f0b083fa6f7ec3b5a80d74 into abd9437e6d5419a497707748eb935e50976c3b7b Zengwei Yao 2025-06-27 11:32:12 +00:00
  • 14d4052b1b
    Merge 72b0aa3fbf7ef9c4295f3a00324483ff496ab280 into abd9437e6d5419a497707748eb935e50976c3b7b Nagendra Goel 2025-06-27 11:32:08 +00:00
  • 556eebaeae
    Merge f4bf9e4505d047decbc1aec2a46c78c9b4aaf608 into abd9437e6d5419a497707748eb935e50976c3b7b Charlie_Tang 2025-06-27 11:32:06 +00:00
  • 742610ca3b
    Merge 5ea4d94ac0920eb84762939b827a9ef45c7723a4 into abd9437e6d5419a497707748eb935e50976c3b7b Guanbo Wang 2025-06-27 11:32:03 +00:00
  • 5222a01dfc
    Merge 83e2b30a224832d738235743ba88a70a7262e361 into abd9437e6d5419a497707748eb935e50976c3b7b GoVivace 2025-06-27 11:32:01 +00:00
  • 65f0097477
    Merge 01be91217b7205a234af4f70879af75fdf9019b1 into abd9437e6d5419a497707748eb935e50976c3b7b Fangjun Kuang 2025-06-27 11:31:33 +00:00
  • 6fbafb0c20
    Merge feb526c2a40e0919a2f4cf9cde9d35eb442696ca into abd9437e6d5419a497707748eb935e50976c3b7b Fangjun Kuang 2025-06-27 11:31:26 +00:00
  • 244379f580
    Merge 2704d589df6ef82ffdb4cb3c504004e1e63b61c7 into abd9437e6d5419a497707748eb935e50976c3b7b Liyong.Guo 2025-06-27 11:31:20 +00:00
  • 09004707d0
    Merge 4319a187b392d2a0fb6c211c96f7b30639eb3234 into abd9437e6d5419a497707748eb935e50976c3b7b Fangjun Kuang 2025-06-27 11:31:18 +00:00
  • 7500b0fbf8
    Merge ec8fa55bcf4894e63b8f8626d6f8a77aaceab82b into abd9437e6d5419a497707748eb935e50976c3b7b Piotr Żelasko 2025-06-27 11:31:10 +00:00
  • 0ac15ce71d
    Merge 65212ee0041db43c826c5331c907730ba7c87cd4 into abd9437e6d5419a497707748eb935e50976c3b7b Fangjun Kuang 2025-06-27 11:31:06 +00:00
  • c78f4c6097
    Merge ca15b32b769e111be808fd58a8c62e286d88e43d into abd9437e6d5419a497707748eb935e50976c3b7b Piotr Żelasko 2025-06-27 11:30:55 +00:00
  • a7591cba68
    Merge 359ffce6c9f1851e2eebc9bdea4e965c226671f0 into abd9437e6d5419a497707748eb935e50976c3b7b Yifan Yang 2025-06-25 19:35:53 +05:30
  • abd9437e6d
    Add more wheels for piper-phonemize (#1969) Fangjun Kuang 2025-06-24 14:49:16 +08:00
  • a879de95a3 deploy: 93940904d8aa837ef9d4def90f481462976dcbd1 csukuangfj 2025-06-24 06:33:27 +00:00
  • 93940904d8 fix windows Fangjun Kuang 2025-06-24 14:32:12 +08:00
  • 896009611f deploy: 5fe1fad4ec53967f4ca339d9b90d7bb18280f93d csukuangfj 2025-06-24 03:18:26 +00:00
  • 5fe1fad4ec update ci Fangjun Kuang 2025-06-24 11:17:43 +08:00
  • bbbf798375 update piper-phonemize wheels Fangjun Kuang 2025-06-24 11:15:38 +08:00
  • e1cf4dbace
    rm zipvoice (#1967) Wei Kang 2025-06-23 19:22:35 +08:00
  • 0c9bd934c2 rm zipvoice pkufool 2025-06-23 19:16:33 +08:00
  • 343b8fa2dc
    Using non strict match in context graph for contextual words (#1952) Wei Kang 2025-06-19 12:27:15 +08:00
  • f80a2ee110
    Decrease num_buckets & remove shuffle_buffer_size (#1955) Wei Kang 2025-06-19 12:26:37 +08:00
  • 3587c4b3b7
    Fix decoding byte bpes tokens to words. (#1966) Wei Kang 2025-06-19 12:26:01 +08:00
  • 2e1a1af049
    ignore decode errors. Wei Kang 2025-06-19 11:30:21 +08:00
  • ba5ffc711e
    Minor fix. Wei Kang 2025-06-19 11:25:48 +08:00
  • 857507795d
    Fix deocding byte bpes tokens to words. Wei Kang 2025-06-19 11:17:38 +08:00
  • 56349001d6
    Merge branch 'k2-fsa:master' into dev/speechllm Yifan Yang 2025-06-18 21:09:44 +08:00
  • 762f965cf7
    [zipvoice] Add requirements.txt and pinyin.txt, remove k2 from pretrained model inference. (#1965) Wei Kang 2025-06-18 18:38:46 +08:00
  • 53111d0e46 fix for multigpu yfyeung 2025-06-18 07:33:15 +00:00
  • dae64dd08d simplify the requirements for pretrained model inference pkufool 2025-06-18 13:51:40 +08:00
  • c197be2c05 simplify the requirements for pretrained model inference pkufool 2025-06-18 13:50:39 +08:00
  • 39d90356fe fix deepspeed config yfyeung 2025-06-18 04:44:10 +00:00
  • c571a88b59
    Merge branch 'k2-fsa:master' into dev/speechllm Yifan Yang 2025-06-18 12:29:27 +08:00
  • 34639d5249 use padding instead of trimming (suggested by @shylockasr) Yifan Yang 2025-06-03 21:45:47 +08:00
  • 05e3094429 refactor branch exchange in cr-ctc (#1954) Zengwei Yao 2025-05-27 12:09:59 +08:00
  • d23bacc23b fix isort pkufool 2025-06-18 12:07:46 +08:00
  • 88c35c5e29 fix flake8 pkufool 2025-06-18 12:00:05 +08:00
  • df382566dc Add requirements.txt and pinyin.txt needed by zipvoice pkufool 2025-06-18 11:49:56 +08:00
  • 06539d2b9d
    Add Zipvoice (#1964) Wei Kang 2025-06-17 20:17:12 +08:00
  • e45da09009 Minor fixes pkufool 2025-06-17 20:02:05 +08:00
  • dc731ea089 minor fixes pkufool 2025-06-17 19:48:38 +08:00
  • 2376ed2117 add emilia data preparation pipeline pkufool 2025-06-17 19:38:46 +08:00
  • 60572c2444 Minor fixes to infer pretrained model pkufool 2025-06-17 16:02:20 +08:00
  • 8c529ebe90
    Merge pull request #3 from zhu-han/zipvoice Wei Kang 2025-06-17 10:29:42 +08:00
  • ecfc36ba9e Update the paper link Han Zhu 2025-06-17 10:03:25 +08:00
  • 9936d726d2 Add ZipVoice Han Zhu 2025-06-16 09:45:34 +08:00
  • 252e5eb2e1 remove unused local scripts Bailey Hirota 2025-06-13 00:49:40 +09:00
  • fe9f975ec2 changes to train script - no need for limiting utterance length here Bailey Hirota 2025-06-13 00:48:37 +09:00
  • e1f140a50e remove commented out codels Bailey Hirota 2025-06-13 00:33:47 +09:00
  • 78d4e50d0f add stage 6 - update cutset paths to prepare Bailey Hirota 2025-06-12 00:21:52 +09:00
  • da75835639 update manifest dir path Bailey Hirota 2025-06-12 00:20:41 +09:00
  • 5a120cbcb3 add step 4: display manifest stats to mls_eng Bailey Hirota 2025-06-11 18:06:08 +09:00
  • 003e94fac2 Update README.md to reflect MLS English dataset Kinan Martin 2025-06-11 09:19:07 +09:00
  • c7c74b8658 Add failsafe for MLS English dev set key alternate name as validation Kinan Martin 2025-06-11 09:18:28 +09:00
  • c8d932b0c2 Parametrize dev and test split sizes. Kinan Martin 2025-06-10 10:11:33 +09:00
  • a6f60de9dd add utility file for creating subsets of mls english. must be fixed to make dev and test splits have matching sizes to reazonspeech Kinan Martin 2025-06-06 11:44:27 +09:00
  • 052fcc3218 add utility file for updating the storage_path of cutsets for use in the multilingual training recipe directory structure Kinan Martin 2025-06-06 11:42:08 +09:00
  • 6255ba5cb2 fix decode script data module usage Kinan Martin 2025-06-06 11:29:29 +09:00
  • 559f9e2def fix repeat bos and pad id root 2025-06-04 10:02:42 +00:00
  • ce894a7ba2 Combined updates. Changed BBPE path structure, changed dataset path structure, added script to update cutset paths. WIP Kinan Martin 2025-06-04 10:12:39 +09:00
  • 80677a55f8 remove stats root 2025-06-03 00:48:39 -07:00
  • 5becf6927d remove concat three items root 2025-06-03 00:18:21 -07:00
  • 4c0396f8f2 support text2speech ultrachat root 2025-06-02 23:16:03 -07:00
  • 0f88a3a6c3 First working example Fangjun Kuang 2025-05-30 15:42:31 +08:00
  • 516696f3e4 Merge remote-tracking branch 'dan/master' into dataset-parallel-augmentation-example Fangjun Kuang 2025-05-29 17:04:50 +08:00
  • 3b52e0cb9e minor fixes Fangjun Kuang 2025-05-29 12:11:56 +08:00
  • dc74705d20 remove cr-loss Fangjun Kuang 2025-05-29 11:49:30 +08:00
  • 9b95c72d19 copy files Fangjun Kuang 2025-05-29 11:45:17 +08:00