mrq 0dc49ef4d5 documentation update while I wait for more audio (between 4 and 8 seconds per utterance) quantize for nvidia/audio-codec-44khz (I was foolish to think I can get something servicable with just 4 seconds max for an utterance)		2025-02-15 17:42:06 -06:00
..
cleanup_dataset.py
deduplicate_librilight_libritts.py
parse_ppp.py
prepare_librilight.py
process_emilia.py	documentation update while I wait for more audio (between 4 and 8 seconds per utterance) quantize for nvidia/audio-codec-44khz (I was foolish to think I can get something servicable with just 4 seconds max for an utterance)	2025-02-15 17:42:06 -06:00
process_libritts.py	agony	2025-02-12 00:18:24 -06:00
process_nscripter.py
process_seed-tts.py
run.sh
setup.sh
train_tokenizer.py