31 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
Michael Hansen	c9d824c0e2	Require sample rate in augment	2026-03-12 10:48:04 -05:00
Michael Hansen	ded9350eaf	Check if file	2025-11-14 10:47:29 -06:00
Michael Hansen	66ec23fa49	Clean up	2025-09-19 16:21:32 -05:00
Michael Hansen	c0d2dd9ea4	Merge pull request #18 from kahrendt/phoneme-input ... Add support for using phonemes as direct input	2025-09-19 15:19:44 -05:00
Michael Hansen	942002d0ee	Merge branch 'master' into phoneme-input	2025-09-19 15:19:31 -05:00
Michael Hansen	395aef7dbc	Merge pull request #17 from kahrendt/mps-support-v3 ... Add support for MPS (Apple Silicon) backend	2025-09-19 15:09:08 -05:00
Kevin Ahrendt	df5799c141	add support for directly inputting phonemes	2025-09-19 14:42:49 -04:00
Kevin Ahrendt	af7ac7aae2	add support for MPS acceleration	2025-09-19 13:18:06 -04:00
Michael Hansen	5e67370ab3	Update changelog	2025-08-29 15:17:12 -05:00
Michael Hansen	4d7e4b390c	Update README	2025-08-29 15:16:20 -05:00
Michael Hansen	4057c1a620	Upgrade to torch 2, piper 1.3	2025-08-29 12:19:25 -05:00
Michael Hansen	9c1019c932	Merge pull request #5 from kahrendt/fix-batch-min-phoneme ... Use phoneme lengths to trim samples	2024-02-27 11:00:29 -06:00
Kevin Ahrendt	213d4d561a	revert removing webrtcvad	2024-02-27 10:01:27 -05:00
Kevin Ahrendt	172d7b5cae	use phoneme lengths to trim	2024-02-24 16:20:39 -05:00
Michael Hansen	315e555f49	Add pad after bos	2024-02-05 14:18:15 -06:00
Michael Hansen	2dbff77c61	Add --min-phoneme-count	2024-02-05 14:07:29 -06:00
Michael Hansen	77d8c0d4b3	Add configs for mls models	2024-02-03 11:05:58 -06:00
Michael Hansen	195e3bd967	Clean up merge	2023-10-06 12:10:43 -05:00
Michael Hansen	595536c630	Merge remote-tracking branch 'dscripka/master'	2023-10-06 12:00:28 -05:00
Michael Hansen	48e54922df	Switch to libritts_r medium model and use piper-phonemize	2023-10-06 11:56:57 -05:00
dscripka	f1988a4d54	Added use of pywebrtc to remove excessive silence from generated clips	2023-09-11 07:56:17 -04:00
dscripka	b104c2412d	Fixed major bug where batch sizes >= 1 would not generate all input texts	2023-09-10 15:11:45 -04:00
dscripka	28979b8964	Added functionality in 'generate_samples' function to attempt to auto-recover from CUDA OOM errors	2023-08-16 21:44:51 -04:00
dscripka	8544faa48c	typo	2023-08-06 11:20:34 -04:00
dscripka	f82afc6596	Added example of import usage to README	2023-08-06 11:19:56 -04:00
dscripka	8b4b20b712	Cleanup and verbose argument	2023-08-06 11:09:39 -04:00
dscripka	a1f84c5ec3	Refactor to allow for usage of generation within Python script, added automatic resampling to 16khz	2023-06-20 07:35:39 -04:00
Michael Hansen	87f249822f	Add --max-speakers	2023-05-17 16:34:27 -05:00
Michael Hansen	b87c930fa1	Add download link	2023-05-17 12:09:42 -05:00
Michael Hansen	d8aa042b96	Add impulses and rest of code	2023-05-17 12:02:34 -05:00
Michael Hansen	6ecc5192d9	Initial commit	2023-05-16 17:06:57 -05:00