modified logic to determine valid voice folders, also allows subdirs within the folder (for example: ./voices/SH/james/ will be named SH/james)

should fix #203

a bunch of shit i had uncommited over the past while pertaining to VALL-E

added VALL-E inference support (very rudimentary, gimped, but it will load a model trained on a config generated through the web UI)

only include auto in the list of models under setting, nothing else

added mixing models (shamelessly inspired from voldy's web ui)

fixes #176

for real this time show those new vall-e metrics

added showing reported training accuracy and eval/validation metrics to graph

now there should be feature parity between trainers

x_lim and y_lim for graph

actually make parsing VALL-E metrics work

I forget

my sanitizer actually did work, it was just batch sizes leading to problems when transcribing

when the sanitizer thingy works in testing but it doesn't outside of testing, and you have to retranscribe for the fourth time today

why does this keep happening to me

more cleanup, use 24KHz for preparing for VALL-E (encodec will resample to 24Khz anyways, makes audio a little nicer), some other things

remove redundant phonemize for vall-e (oops), quantize all files and then phonemize all files for cope optimization, load alignment model once instead of for every transcription (speedup with whisperx)

do not write current whisper.json if there's no changes

added whisper transcription 'sanitizing' (collapse very short transcriptions to the previous segment) (I really have to stop having several copies spanning several machines for AIVC, I keep reverting shit)

disable diarization for whisperx as it's just a useless performance hit (I don't have anything that's multispeaker within the same audio file at the moment)

ugh

now whisperx should output json that aligns with what's expected

fixes for whisperx batching

begrudgingly added back whisperx integration (VAD/Diarization testing, I really, really need accurate timestamps before dumping mondo amounts of time on training a dataset)

Fixed #167

cleanups, realigning vall-e training

ugh

Added option to unsqueeze sample batches after sampling

forgot to not require it to be relative

option to set results folder location

tweaks

brain worms

yammed

yammed

cleanup, metrics are grabbed for vall-e trainer

forgot to separate phonemes by spaces for [redacted]

cleaned up some prepare dataset code

setup bnb on windows as needed

unk hunting

fixes v2

fixed reload tts being broken due to being as untouched as I am

yammed

fixes

ops

ops

I forgot I undid the thing in DLAS

oops

preparations for training an IPA-based finetune

fixed not cleaning up states while training (oops)