Commit Graph

62 Commits

Author SHA1 Message Date
unknown
d5c307b56a add logger 2024-10-29 17:30:28 +02:00
unknown
8a5041ef9f update 2024-10-29 17:14:03 +02:00
unknown
886500ac97 update 2024-10-29 17:02:21 +02:00
unknown
2ca1fb7c25 update 2024-10-29 16:46:38 +02:00
unknown
3409192662 update 2024-10-29 16:44:23 +02:00
unknown
37eb3b50da add tensorboard and add export sample for mel and audio 2024-10-29 14:20:50 +02:00
Jarod Mica
d601a70ad1 Revert "."
This reverts commit 5089dfe51b.
2024-10-29 03:25:02 -07:00
Jarod Mica
5089dfe51b . 2024-10-29 09:48:17 +00:00
SWivid
551857b268 minor fix. 2024-10-29 13:40:40 +08:00
SWivid
91881841dd fix. wider search for silence 2024-10-29 13:26:55 +08:00
SWivid
82d04be12a Update infer README.md 2024-10-29 13:18:04 +08:00
SWivid
85089a276b fix. change 18s threshold to 15s for clipping 2024-10-29 13:11:29 +08:00
SWivid
456456971b fix. proper clip for long infer ref_audio if no silence found 2024-10-29 12:50:55 +08:00
unknown
2dddb10c36 fix vocab file take from the project 2024-10-28 21:09:45 +02:00
unknown
41eb33c5c6 add stream output 2024-10-28 12:51:51 +02:00
unknown
5427f28a6d fix wrong value print vocab 2024-10-28 10:26:19 +02:00
unknown
9db5de651b add note about ema 2024-10-27 21:10:26 +02:00
unknown
8c3810a66c change name make more clear the preetain need path 2024-10-27 20:48:45 +02:00
unknown
eb19d9d928 fix path 2024-10-27 20:39:04 +02:00
unknown
3af98f2a52 add settings 2024-10-27 20:21:12 +02:00
unknown
48e3eb1c57 add settings 2024-10-27 20:19:59 +02:00
unknown
0f2a9230ec add settings 2024-10-27 20:16:21 +02:00
unknown
2eae16b4a3 Do not overwrite the vocab if it already exists ! 2024-10-27 19:30:04 +02:00
unknown
0de2e531d4 fix extend 2024-10-27 19:14:04 +02:00
unknown
e1e3b26987 fix space curse problem with utf-8-sig 2024-10-27 18:01:46 +02:00
unknown
d864425144 add use_ema in model test 2024-10-27 15:50:14 +02:00
lpscr
3941e8102f make happy all other language dont suport the symbols in vocab , now you can finetune by extend (#293)
* fix path

* change name

* change name

* fix path

* fix last per steps and add notes

* change order tab add note in vocab check tab

* add note in reduse checkpoint tab

* note in reduse checkpoint tab update

* extend vocab to train language miss symbols

* change enten to ,

* hide the option create new vocab , change order tab , add some info
2024-10-27 21:03:22 +08:00
jpgallegoar
2056f5de41 Format 2024-10-26 19:11:55 +02:00
jpgallegoar
a8bb34bde2 Format 2024-10-26 19:00:07 +02:00
jpgallegoar
4eb305418d Added Insert button on Multi-Speech tab 2024-10-26 18:58:43 +02:00
lpscr
c1a9986a15 gradio finetune fix last per step and add note (#284)
* fix path

* change name

* change name

* fix path

* fix last per steps and add notes

* change order tab add note in vocab check tab

* add note in reduse checkpoint tab

* note in reduse checkpoint tab update
2024-10-26 17:57:04 +08:00
Justin John
be090b0d0a Cleanup 2024-10-26 10:48:48 +05:30
Justin John
ed179067df Add speed control to F5-TTS inference CLI
- Added support for the --speed argument to control the speed of generated audio.
- Updated the CLI to accept a speed parameter with a default value of 1.0.
- Adjusted the infer_process to apply the speed factor during TTS generation.
2024-10-26 10:45:11 +05:30
Justin John
bfcb73688e fix path in finetune_gradio for finetune-cli 2024-10-26 07:16:32 +05:30
SWivid
75d1ceba56 fix auto-reset microphone 2024-10-26 04:32:56 +08:00
SWivid
29b5a4784f auto revert to microphone ready 2024-10-26 04:04:44 +08:00
SWivid
7f826e0261 minor fix for infer_gradio.py 2024-10-26 03:49:34 +08:00
SWivid
f93a9ebfaf Update infer_gradio.py 2024-10-26 03:36:35 +08:00
SWivid
cc5ded275c fix space demo 2024-10-26 03:15:34 +08:00
SWivid
54d557789e fix add @gpu_decorator for space demo 2024-10-26 02:59:19 +08:00
lpscr
0676df5982 fix path in finetune-cli working with new update (#270)
* fix path
* change name
2024-10-26 01:52:55 +08:00
jpgallegoar
984933f155 Fixed formatting 2024-10-25 14:23:12 +02:00
jpgallegoar
17080078d3 Improved Multi-Speech 2024-10-25 14:20:21 +02:00
SWivid
76f56979ff merge podcast and multistyle 2024-10-25 17:49:18 +08:00
lpscr
33d903be45 fix path finetune gradio (#262)
* fix path
* change name
2024-10-25 16:50:55 +08:00
SWivid
f69a60287b finish infer dependencies; update readmes 2024-10-25 03:48:53 +08:00
SWivid
adca73b4d0 fix rel path for pip pkg usage 2024-10-25 00:58:29 +08:00
SWivid
041c3391d2 fix vocab path for pip pkg infer 2024-10-25 00:42:01 +08:00
SWivid
9f5328d95d update infer_gradio credit 2024-10-25 00:16:24 +08:00
SWivid
8629c6f91f initial updates for infer stuffs 2024-10-24 23:51:20 +08:00