ben_mkiv
  • Joined on Aug 16, 2023
Loading Heatmap…

ben_mkiv commented on issue mrq/ai-voice-cloning#349

unload_tts() doesnt unload the voice model from video memory

i tried to find the root of the problem, but haven't succeed yet. thought you might have an idea. is run_training() maybe already called on another thread/process and therefor not able to…

2023-08-26 15:50:33 +07:00

ben_mkiv created pull request mrq/ai-voice-cloning#350

Websocket fixes / additions

2023-08-26 15:46:43 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • 13b65d8775 Merge branch 'master' of https://git.ecker.tech/ben_mkiv/ai-voice-cloning
  • b72f2216bf added websocket server arguments to enabled it (now disabled by default) and to specify the address/port to listen on
  • 6f0f148782 websocket server: fix for model loading (just overriding args didn't do it after all...)
  • b4dc103931 I don't know how I did not commit the 'sample from the voices to construct the input prompt for vall-e' change but this helps
  • a657623cbc updated vall-e training template to use path-based speakers because it would just have a batch/epoch size of 1 otherwise; revert hardcoded 'spit processed dataset to this path' from my training rig to spit it out in a sane spot
  • Compare 14 commits »

2023-08-26 15:44:18 +07:00

ben_mkiv opened issue mrq/ai-voice-cloning#349

unload_tts() doesnt unload the voice model from video memory

2023-08-26 14:46:52 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • 578a5bcadd websocket server: fix for model loading (just overriding args didn't do it after all...)

2023-08-25 23:41:40 +07:00

ben_mkiv created pull request mrq/ai-voice-cloning#341

fix filename generation which didn't work and overwrote existing files

2023-08-24 08:05:31 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • 00b173857d fix filename generation which didn't work and overwrote existing files

2023-08-24 07:58:23 +07:00

ben_mkiv commented on pull request mrq/ai-voice-cloning#336

favor existing arguments from parameters (kwargs) over global (args)

thank you, i've reduced the PR to overriding the args parameters in the websocket server, ignore the commit spam... squashed it down to one commit

2023-08-23 17:42:42 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • ce24ba41e2 Websocket server, override args parameters for model settings (squashed)

2023-08-23 17:41:21 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

2023-08-23 17:29:40 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

2023-08-23 17:24:45 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • a9d2f241a5 Revert "favor existing arguments from parameters (kwargs) over global (args)"

2023-08-23 17:23:33 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • a17078a792 Revert "added parameter to specify the autoregressive_model (tho it still loads the default model first, and then loads the target model, which seems to be because TTS loading just loads whatever is set in the settings first)"

2023-08-23 17:21:54 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • f087b45943 Revert "added parameter to specify the autoregressive_model (tho it still loads the default model first, and then loads the target model, which seems to be because TTS loading just loads whatever is set in the settings first)"

2023-08-23 17:11:53 +07:00

ben_mkiv created pull request mrq/ai-voice-cloning#336

favor existing arguments from parameters (kwargs) over global (args)

2023-08-22 21:41:47 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • d1dbe3e464 added parameter to specify the autoregressive_model (tho it still loads the default model first, and then loads the target model, which seems to be because TTS loading just loads whatever is set in the settings first)

2023-08-22 21:35:44 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • 89102347a9 favor existing arguments from parameters (kwargs) over global (args)

2023-08-22 21:12:21 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • 5d73d9e71c small QoL change to the StringNone helper, to allow generated text to be "None", maybe someone wants to generate that, we never know...

2023-08-22 19:54:07 +07:00

ben_mkiv created pull request mrq/ai-voice-cloning#334

websocket server: API change(!), better response format

2023-08-22 19:42:01 +07:00

ben_mkiv pushed to master at ben_mkiv/ai-voice-cloning

  • 9abcb0f193 websocket server: API change(!), better response format

2023-08-22 19:38:41 +07:00