Commit graph

430 commits

Author SHA1 Message Date
Joseph Cheek
c16ff335e4 Fix test_initial_state() test failure on apple silicon. 2025-07-26 17:07:25 -06:00
Fireblade2534
29066f7c9f
Merge pull request #354 from MiggiV2/master
Update PyTorch to 2.7.1+cu128 for RTX 50 Series GPU Support
2025-07-04 18:50:59 -04:00
Fireblade2534
f8c89161f6
Merge pull request #350 from fireblade2534/master
Fix custom phenomes and make them much more robust
2025-06-27 18:34:26 -04:00
Miggi
8a55cd5bf5 Update torch to 2.7.1 & Cuda 12.8.1 in Docker 2025-06-27 14:30:47 +02:00
Fireblade2534
cab29016af
Merge branch 'remsky:master' into master 2025-06-25 20:16:41 -04:00
Fireblade2534
2a6d2ae483 Fix custom phenomes and make them more robust 2025-06-26 00:16:06 +00:00
Fireblade2534
6b1e9d9128
Merge pull request #339 from fireblade2534/master
Release
2025-06-18 18:03:18 -04:00
Fireblade2534
ac491a9b18 Release 2025-06-18 22:02:33 +00:00
Fireblade2534
223d3077c2
Merge pull request #316 from JCallicoat/volume_setting
Add a volume multiplier setting
2025-06-16 17:10:17 -04:00
Fireblade2534
ec947a7b54
Merge branch 'master' into volume_setting 2025-06-16 12:43:26 -04:00
Fireblade2534
cd82dd0735 Added a volume multiplier as a request parameter 2025-06-16 16:39:30 +00:00
remsky
1e15433d24
Merge pull request #311 from mbailey/master 2025-06-13 07:58:14 -06:00
Fireblade2534
d7d90cdc9d simplified some normalization and added more tests 2025-06-12 16:00:06 +00:00
Fireblade2534
dbb66ff1e1
Merge pull request #322 from EasyMetaAu/up-master
Improve Audio Pause Handling, MP3 Encoding, and Robust Text Normalization/Splitting
2025-06-12 11:25:06 -04:00
Lukin
f7fb9c524a Refactor TTS service tests: Update test_get_voice_path_combined to verify path format for combined voices, removing mock for os.path.join and enhancing assertions for path validation. 2025-06-01 10:35:59 +08:00
Lukin
0b2260602a Update TTS service tests: Enhance test_get_voice_path_combined by mocking os.path.join to ensure correct path generation for combined voices. 2025-06-01 10:28:35 +08:00
Lukin
888e3121ff Refactor text normalization: Move handling of problematic symbols to occur after number and money processing to improve accuracy in text normalization. 2025-06-01 10:18:24 +08:00
Lukin
84d2a4d806 Enhance TTS text processing: Implement pause tag handling in smart_split, allowing for better audio chunk generation with pauses. Update related tests to validate new functionality and ensure compatibility with existing features. 2025-05-30 23:06:41 +08:00
Lukin
ab8ab7d749 Refactor audio processing and text normalization: Update audio normalization to use absolute amplitude threshold, enhance streaming audio writer with MP3 container options, and improve text normalization by stripping spaces and handling special characters to prevent audio artifacts. 2025-05-30 22:52:58 +08:00
Fireblade2534
543cbecc1a
Merge pull request #321 from jiaohuix/master
feat(text): add Chinese punctuation-based sentence splitting for bett…
2025-05-28 10:57:55 -04:00
Fireblade2534
b89da1ff28 Make the code cleaner and add tests 2025-05-28 14:53:00 +00:00
jiaohuix
9c279f2b5e feat(text): add Chinese punctuation-based sentence splitting for better TTS 2025-05-26 15:30:03 +08:00
Fireblade2534
ce22f60344
Fix readme claiming that Korean is supported 2025-05-24 21:31:38 -04:00
JCallicoat
75963c4aeb Add a volume multiplier setting
Allow configuring output volume via multiplier applied to np array of
audio chunk.

Defaults to 1.0 which is no-op.

Fixes #110
2025-05-22 07:02:34 -05:00
Mike Bailey
b0f46276eb
Update paths.py
Use root_dir instead of /app

This was breaking things for me when I started the app from a script that was not in the root_dir
2025-05-11 01:00:41 +10:00
Fireblade2534
6f1450c5b4
Merge pull request #310 from fireblade2534/master
Some checks failed
CI / test (3.10) (push) Has been cancelled
Added some better saftey checks to captioned speech
2025-05-09 09:13:02 -04:00
Fireblade2534
243d98e339
Merge branch 'remsky:master' into master 2025-05-09 09:08:55 -04:00
Fireblade2534
1ac9924c3e Added some better safty checks to captioned speech 2025-05-09 13:04:53 +00:00
Fireblade2534
36197b9266
Merge pull request #304 from fireblade2534/Simple-phenome-fixes
Fixed phenomes
2025-05-05 22:03:52 -04:00
Fireblade2534
15bb614c07
Merge pull request #303 from fireblade2534/Fixing-timestamps
Fixing normalization
2025-05-05 22:03:42 -04:00
Fireblade2534
acee979225 Fixed phenomes 2025-05-02 13:34:08 +00:00
Fireblade2534
c80d4feb69 Fixed formatting 2025-05-01 18:16:36 +00:00
Fireblade2534
f89b76d5d6 Added more tests 2025-05-01 18:02:35 +00:00
Fireblade2534
48e096b441 Massivly improved how number, time, and money normalization 2025-05-01 17:53:01 +00:00
remsky
2c7e1e8c0b
Merge pull request #291 from RigleGit/patch-1 2025-04-22 09:51:16 -06:00
Rodrigo Iglesias
83325b749d
Update Dockerfile to install Rust
Rust is required to build sudachipy and pyopenjtalk-plus
2025-04-17 16:07:15 +02:00
remsky
f1fa340494
Update README.md
Some checks failed
CI / test (3.10) (push) Has been cancelled
2025-04-05 04:09:28 -06:00
remsky
d826de87ed Minor change to trigger new release 2025-04-04 22:07:26 -06:00
remsky
59be119cc4
Merge pull request #276 from remsky/maintenance/automations
Maintenance/automations
2025-04-04 21:19:09 -06:00
remsky
10f240daad Add tag existence check in release workflow 2025-04-04 19:38:27 -06:00
remsky
10caafe3fb Adjust CONTRIBUTING.md, readme docker information and notes 2025-04-04 19:30:13 -06:00
remsky
ba796ef003 Update CHANGELOG.md for version 0.3.0: add new features, changes, fixes, and removals 2025-04-04 18:53:13 -06:00
remsky
a28de9bc66
Delete readme-parts/config.yaml 2025-04-04 17:28:06 -06:00
Fireblade2534
d004b6d304
Apply suggestions from copilot
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-04-04 19:06:59 -04:00
Fireblade2534
7f0e06ea6b
Update normalizer.py 2025-04-04 19:06:13 -04:00
remsky
afa879546c CONTRIBUTING + Ruff format 2025-04-04 16:58:07 -06:00
remsky
447f9d360c Ruff check 2025-04-04 16:50:46 -06:00
remsky
6a2d3a54cf Bump version to 0.3.0 and update related configurations; add misaki patch script and remove obsolete build workflow 2025-04-04 16:49:10 -06:00
remsky
e2313abe72
Merge pull request #266 from rampadc/master
Some checks failed
CI / test (3.10) (push) Has been cancelled
start-gpu_mac.sh: removed duplicated env and align with other shell scripts
2025-04-02 22:04:04 -06:00
Cong Nguyen
64ced408b7 removed duplicated env and align with other shell scripts 2025-03-31 03:54:41 +11:00