db317c358b
feat: add language-aware sentence tokenization ( #1269 )
...
* feat: add language-aware sentence tokenization
* feat: add missing punkt languages
---------
Co-authored-by: pulkit <129310466+p1kit@users.noreply.github.com >
Co-authored-by: Barabazs <31799121+Barabazs@users.noreply.github.com >
2025-10-21 15:57:26 +02:00
0fa81b31f1
feat: add Swedish alignment model ( #1110 )
...
---------
Co-authored-by: Paffe <paffe@Sigge.home >
2025-10-15 08:08:18 +02:00
a51ae7a81a
feat: add centralized logging to replace ad-hoc print statements ( #1254 )
...
* feat: add logging utility functions
* feat: add logging setup and log level argument to CLI
* feat: integrate logging across modules
2025-10-10 08:41:06 +02:00
3b1b9a8c4d
refactor: rename types.py to schema.py to avoid stdlib conflict
2025-10-09 14:25:58 +02:00
b1c8ac7de6
Change alignment model for Vietnamese language
...
Since the current model is a wav2vec2 pre-trained model for Vietnamese audio, it won't work with alignment tasks. To make it work as expected, I recommend chaining to a fine-tuned ASR version.
2025-10-03 09:41:03 +02:00
bf150e442e
feat: update Punkt tokenizer to use pre-trained model and handle missing data
2025-10-03 09:11:24 +02:00
c7d31883bc
Add jr, sr, and ph.d to punkt abbreviations
2025-10-01 08:59:53 +02:00
e0833da5dc
Fix: Ensure integer tensor indexing in get_wildcard_emission()
2025-06-27 09:17:44 +02:00
e7712f496e
refactor: update import statements to use explicit module paths across multiple files
2025-03-25 16:24:21 +01:00
8c58c54635
Revert "feat: add Basque alignment model ( #1074 )" ( #1077 )
...
This reverts commit 0d9807adc5
2025-03-05 15:19:23 +01:00
0d9807adc5
feat: add Basque alignment model ( #1074 )
2025-03-04 14:55:30 +01:00
4db839018c
feat: add Tagalog (tl - Filipino) Phoneme-based ASR Model ( #1067 )
2025-02-23 09:59:48 +01:00
36d2622e27
feat: add Latvian align model
2025-01-25 09:45:17 +01:00
ffbc73664c
change the docstrings and comments to English
2025-01-13 22:56:48 +08:00
289eadfc76
fix a merge error.
2025-01-13 20:26:27 +08:00
22a93f2932
Merge branch 'main' into main
2025-01-13 19:34:21 +08:00
2f93e029c7
feat: add SegmentData type for temporary processing during alignment
2025-01-13 10:45:50 +01:00
024bc8481b
refactor: consolidate segment data handling in alignment function
2025-01-13 10:45:50 +01:00
f286e7f3de
refactor: improve type hints and clean up imports
2025-01-13 10:45:50 +01:00
4ebfb078c5
make no beam consistent with backtrack.
2025-01-09 23:13:11 +08:00
65b2332e13
make align a bit faster.
2025-01-09 19:33:26 +08:00
69281f3a29
support timestamps for numbers.
2025-01-09 15:23:40 +08:00
9a8967f27e
refactor: add type hints
2025-01-05 11:48:24 +01:00
1c6d9327bc
feat: use model_dir as cache_dir for wav2vec2 ( #681 )
2025-01-01 13:22:27 +01:00
9b9e03c4cc
feat: update Norwegian models ( #687 )
...
Updated Norwegian Bokmål and Norwegian Nynorsk models
Co-authored-by: Barabazs <31799121+Barabazs@users.noreply.github.com >
2024-12-16 11:08:48 +01:00
19eff8e79a
feat: add new align models ( #922 )
...
Co-authored-by: Barabazs <31799121+Barabazs@users.noreply.github.com >
2024-12-16 11:06:43 +01:00
6f3bc5b7b8
Added Romanian phoneme-based ASR model ( #791 )
...
Co-authored-by: Barabazs <31799121+Barabazs@users.noreply.github.com >
2024-12-16 08:09:53 +01:00
3f339f9515
Update alignment.py - remove commented-out alignment modules for hr language
2024-08-09 13:00:12 +02:00
9a9b6171e6
Update alignment.py - trying another hr alignment
2024-08-08 08:37:55 +02:00
59b4d88d1d
Update alignment.py - trying another hr alignment file
2024-08-08 08:29:11 +02:00
6f70aa6beb
Update alignment.py - added croatian (hr) language
2024-08-08 08:10:55 +02:00
912920c591
Update alignment.py - added alignment for sk and sl languages
2024-08-07 10:05:17 +02:00
5dfbfcbdc0
Adding Norwegian Bokmål and Norwegian Nynorsk
...
Adding Wav2Vec2-models for Norwegian Bokmål and Norwegian Nynorsk. The models are testet together with WhisperX, and works great. For Bokmål I have added the 1B model, even if I see fairly little difference between that and the 300M model. For Norwegian Nynorsk only a 300M exist.The quality of the Wav2Vec models are also reported here: https://arxiv.org/abs/2307.01672
2023-12-19 08:48:21 +01:00
5756b0fb13
Update alignment.py
2023-11-17 05:21:23 +05:30
aaaa3de810
Update alignment.py
2023-11-17 05:18:19 +05:30
9f41c49fe5
Add align model for catalan language.
2023-11-16 11:43:36 +01:00
0c7f32f55c
no align based on space
2023-11-03 19:47:00 +01:00
b69956d725
.
2023-10-16 20:43:37 +03:00
02c0323777
fix
2023-10-15 16:25:15 +03:00
8049dba2f7
fix minimum input length for torch wav2vec2 models
2023-10-06 00:41:23 +03:00
a0b6459c8b
fix: ZeroDivisionError when --print_progress True
2023-09-27 20:10:43 +05:30
f137f31de6
Update alignment.py
2023-09-25 15:33:06 +09:00
4e28492dbd
Update alignment.py
2023-08-17 14:57:53 +02:00
6cb7267dc2
Update alignment.py
2023-08-17 14:56:54 +02:00
abbb66b58e
Update alignment.py
2023-08-17 14:53:53 +02:00
65688208c9
Update alignment.py
2023-08-16 16:18:00 +02:00
1bb4839b0f
Update alignment.py
2023-08-16 16:13:28 +02:00
14e593f60b
Update alignment.py
2023-08-16 16:08:25 +02:00
18d5fdc995
Add telugu language to alignment.py
2023-08-10 12:13:52 +01:00
423667f00b
Update alignment.py
2023-08-09 17:08:56 +05:30
pplkit
Paffe
Barabazs
Nguyen Binh
Alex Cannan
Howard
Xabi
Amerogin Kamid
Reinis Ivanovs
liupeng
bfs18
bnitsan
pere
Ismael Ruiz Ranz
Ruhollah Majdoddin
Jan Panoch
Kurian Benoy
David Martin Rius
Simon Sorg
Mahmoud Ashraf
Manohar Reddy
piuy11
awerks
Max Bain
Ayushi-Desynova