mrq

18 Followers · 0 Following

• https://git.ecker.tech/ aims to provide a place to share my efforts while maintaining true ownership of my code, as I do not trust GitHub.

  XMR: 4B9TQdkAkBFYrbj5ztvTx89e5LpucPeTSPzemCihdDi9EBnx7btn8RDNZTBz2zihWsjMnDkzn5As1LU6gLv3KQy8BLsZ8SG

• Joined on 2022-10-10

Repositories

15

Projects Packages Public Activity Starred Repositories

mrq pushed to master at mrq/vall-e 2024-10-11 00:36:05 +00:00

85d85c1351 more arg creep for demo page

mrq pushed to master at mrq/vall-e 2024-10-11 00:09:56 +00:00

301468f519 <<

mrq pushed to master at mrq/vall-e 2024-10-11 00:00:20 +00:00

75a4c866d6 more demo page tweaks, added arg to force enable/disable LoRAs for inferencing (to-do: setup arg flags to handle this, and checkbox in web UI)

mrq pushed to master at mrq/vall-e 2024-10-10 23:59:47 +00:00

ac3d23a905 more demo page tweaks, added arg to force enable/disable LoRAs for inferencing (to-do: setup arg flags to handle this, and checkbox in web UI)

mrq pushed to master at mrq/vall-e 2024-10-10 23:44:39 +00:00

2053ea7c95 ugh

mrq pushed to master at mrq/vall-e 2024-10-10 23:39:05 +00:00

ad58adc41a more demo page tweaks, added arg to force enable/disable LoRAs for inferencing (to-do: setup arg flags to handle this, and checkbox in web UI)

mrq pushed to master at mrq/vall-e 2024-10-10 18:48:49 +00:00

96d05be73c demo page tweaks

mrq pushed to master at mrq/vall-e 2024-10-10 18:36:32 +00:00

2ea978f318 added --eval-random-text-prompts to use random text prompts for eval pass, added --random-prompts for demo page and --lora to use a sample with the lora disabled, probably finally fixed validation dataloader breaking on eval

mrq pushed to master at mrq/vall-e 2024-10-09 00:56:39 +00:00

52299127ab fix vall_e.emb.process

mrq pushed to master at mrq/vall-e 2024-10-09 00:20:53 +00:00

0656a762af fix vall_e.emb.transcriber

mrq pushed to master at mrq/vall-e 2024-10-06 03:50:15 +00:00

acdce66d4e readme tweaks, set the (unused) default model download URL back to the base ar+nar-llama-8 model, as ar+nar-tts+stt-llama-8 was renamed back to it since it performs well

mrq pushed to master at mrq/vall-e 2024-10-05 03:26:48 +00:00

84c7419001 faster

mrq pushed to master at mrq/vall-e 2024-10-05 03:15:38 +00:00

a507b769a1 sped up inferencing by not doing .tolist() for rep pen / length pen (and a bug fix in the web UI from prev commit)

mrq pushed to master at mrq/vall-e 2024-10-04 23:53:25 +00:00

4a8e3ccf06 README tweaks, added --input-prompt-prefix as an experiment (its literally better to just not do this, but i'll retain it in case i have a revelation on how to improve it)

a9fa0898a9 tweaked demo page script to sample speakers instead

Compare 2 commits »

mrq pushed to master at mrq/vall-e 2024-09-28 14:45:59 +00:00

2f1dca3089 added language selection in web UI, tweaked demo script

mrq pushed to master at mrq/vall-e 2024-09-28 01:24:06 +00:00

10df2ef5f3 fixed oversight where input audio does not resample (lol...)

mrq pushed to master at mrq/vall-e 2024-09-26 23:53:04 +00:00

039482a48e don't do eval on stt because it's so slow and I don't even bother doing any metrics against it anyways (to-do: make this a flag)

mrq pushed to master at mrq/vall-e 2024-09-26 23:50:00 +00:00

73ac6e4036 don't do eval on stt because it's so slow and I don't even bother doing any metrics against it anyways (to-do: make this a flag)

mrq pushed to master at mrq/vall-e 2024-09-26 23:34:03 +00:00

ff7a1b4163 coerce into path for other sampler_types (it's required for sampling for similar utterances)

mrq pushed to master at mrq/vall-e 2024-09-26 21:22:49 +00:00

f24547ad4e add top_k sampling / offset for prompt similar utterance sampling