Commit Graph

1249 Commits

Author SHA1 Message Date
James Betker
e9bb692490 fixed aligned_latent 2022-05-06 00:20:21 -06:00
James Betker
1609101a42 musical gap filler 2022-05-05 16:47:08 -06:00
James Betker
d66ab2d28c Remove unused waveform_gens 2022-05-04 21:06:54 -06:00
James Betker
47662b9ec5 some random crap 2022-05-04 20:29:23 -06:00
James Betker
c42c53e75a Add a trainable network for converting a normal distribution into a latent space 2022-05-02 09:47:30 -06:00
James Betker
ab219fbefb output variance 2022-05-02 00:10:33 -06:00
James Betker
3b074aac34 add checkpointing 2022-05-02 00:07:42 -06:00
James Betker
ae5f934ea1 diffwave 2022-05-02 00:05:04 -06:00
James Betker
b712d3b72b break out get_conditioning_latent from unified_voice 2022-05-01 23:04:44 -06:00
James Betker
afa2df57c9 gen3 2022-04-30 10:41:38 -06:00
James Betker
8aa6651fc7 fix surrogate loss return in waveform_gen2 2022-04-28 10:10:11 -06:00
James Betker
f02b01bd9d reverse univnet classifier 2022-04-20 21:37:55 -06:00
James Betker
9df85c902e New gen2
Which is basically a autoencoder with a giant diffusion appendage attached
2022-04-20 21:37:34 -06:00
James Betker
b4549eed9f uv2 fix 2022-04-20 00:27:38 -06:00
James Betker
24fdafd855 fix2 2022-04-20 00:03:29 -06:00
James Betker
0af0051399 fix 2022-04-20 00:01:57 -06:00
James Betker
419f4d37bd gen2 music 2022-04-19 23:38:37 -06:00
James Betker
8fe0dff33c support tts typing 2022-04-16 23:36:57 -06:00
James Betker
48cb6a5abd misc 2022-04-16 20:28:04 -06:00
James Betker
147478a148 cvvp 2022-04-16 20:27:46 -06:00
James Betker
546ecd5aeb music! 2022-04-15 21:21:37 -06:00
James Betker
254357724d gradprop 2022-04-15 09:37:20 -06:00
James Betker
fbf1f4f637 update 2022-04-15 09:34:44 -06:00
James Betker
82aad335ba add distributued logic for loss 2022-04-15 09:31:48 -06:00
James Betker
efe12cb816 Update clvp to add masking probabilities in conditioning and to support code inputs 2022-04-15 09:11:23 -06:00
James Betker
8ea5c307fb Fixes for training the diffusion model on autoregressive inputs 2022-04-11 11:02:44 -06:00
James Betker
a3622462c1 Change latent_conditioner back 2022-04-11 09:00:13 -06:00
James Betker
03d0b90bda fixes 2022-04-10 21:02:12 -06:00
James Betker
19ca5b26c1 Remove flat0 and move it into flat 2022-04-10 21:01:59 -06:00
James Betker
81c952a00a undo relative 2022-04-08 16:32:52 -06:00
James Betker
944b4c3335 more undos 2022-04-08 16:31:08 -06:00
James Betker
032983e2ed fix bug and allow position encodings to be trained separately from the rest of the model 2022-04-08 16:26:01 -06:00
James Betker
09ab1aa9bc revert rotary embeddings work
I'm not really sure that this is going to work. I'd rather explore re-using what I've already trained
2022-04-08 16:18:35 -06:00
James Betker
2fb9ffb0aa Align autoregressive text using start and stop tokens 2022-04-08 09:41:59 -06:00
James Betker
423293e518 fix xtransformers bug 2022-04-08 09:12:46 -06:00
James Betker
048f6f729a remove lightweight_gan 2022-04-07 23:12:08 -07:00
James Betker
e634996a9c autoregressive_codegen: support key_value caching for faster inference 2022-04-07 23:08:46 -07:00
James Betker
d05e162f95 reformat x_transformers 2022-04-07 23:08:03 -07:00
James Betker
7c578eb59b Fix inference in new autoregressive_codegen 2022-04-07 21:22:46 -06:00
James Betker
3f8d7955ef unified_voice with rotary embeddings 2022-04-07 20:11:14 -06:00
James Betker
573e5552b9 CLVP v1 2022-04-07 20:10:57 -06:00
James Betker
71b73db044 clean up 2022-04-07 11:34:10 -06:00
James Betker
6fc4f49e86 some dumb stuff 2022-04-07 11:32:34 -06:00
James Betker
305dc95e4b cg2 2022-04-06 21:24:36 -06:00
James Betker
e011166dd6 autoregressive_codegen r3 2022-04-06 21:04:23 -06:00
James Betker
33ef17e9e5 fix context 2022-04-06 00:45:42 -06:00
James Betker
37bdfe82b2 Modify x_transformers to do checkpointing and use relative positional biases 2022-04-06 00:35:29 -06:00
James Betker
09879b434d bring in x_transformers 2022-04-06 00:21:58 -06:00
James Betker
cdd12ff46c Add code validation to autoregressive_codegen 2022-04-04 09:51:41 -06:00
James Betker
99de63a922 man I'm really on it tonight.... 2022-04-02 22:01:33 -06:00
James Betker
a4bdc80933 moikmadsf 2022-04-02 21:59:50 -06:00
James Betker
1cf20b7337 sdfds 2022-04-02 21:58:09 -06:00
James Betker
b6afc4d542 dsfa 2022-04-02 21:57:00 -06:00
James Betker
4c6bdfc9e2 get rid of relative position embeddings, which do not work with DDP & checkpointing 2022-04-02 21:55:32 -06:00
James Betker
b6d62aca5d add inference model on top of codegen 2022-04-02 21:25:10 -06:00
James Betker
2b6ff09225 autoregressive_codegen v1 2022-04-02 15:07:39 -06:00
James Betker
00767219fc undo latent converter change 2022-04-01 20:46:27 -06:00
James Betker
55c86e02c7 Flat fix 2022-04-01 19:13:33 -06:00
James Betker
8623c51902 fix bug 2022-04-01 16:11:34 -06:00
James Betker
f6a8b0a5ca prep flat0 for feeding from autoregressive_latent_converter 2022-04-01 15:53:45 -06:00
James Betker
3e97abc8a9 update flat0 to break out timestep-independent inference steps 2022-04-01 14:38:53 -06:00
James Betker
a6181a489b Fix loss gapping caused by poor gradients into mel_pred 2022-03-26 22:49:14 -06:00
James Betker
1feade23ff support x-transformers in text_voice_clip and support relative positional embeddings 2022-03-26 22:48:10 -06:00
James Betker
6909f196b4 make code pred returns optional 2022-03-26 08:33:30 -06:00
James Betker
2a29a71c37 attempt to force meaningful codes by adding a surrogate loss 2022-03-26 08:31:40 -06:00
James Betker
45804177b8 more stuff 2022-03-25 00:03:18 -06:00
James Betker
d4218d8443 mods 2022-03-24 23:31:20 -06:00
James Betker
a15970dd97 disable checkpointing in conditioning encoder 2022-03-24 11:49:04 -06:00
James Betker
cc5fc91562 flat0 work 2022-03-24 11:46:53 -06:00
James Betker
b0d2827fad flat0 2022-03-24 11:30:40 -06:00
James Betker
8707a3e0c3 drop full layers in layerdrop, not half layers 2022-03-23 17:15:08 -06:00
James Betker
57da6d0ddf more simplifications 2022-03-22 11:46:03 -06:00
James Betker
f3f391b372 undo sandwich 2022-03-22 11:43:24 -06:00
James Betker
927731f3b4 tts9: fix position embeddings snafu 2022-03-22 11:41:32 -06:00
James Betker
536511fc4b unified_voice: relative position encodings 2022-03-22 11:41:13 -06:00
James Betker
5405ce4363 fix flat 2022-03-22 11:39:39 -06:00
James Betker
e47a759ed8 ....... 2022-03-21 17:22:35 -06:00
James Betker
cc4c9faf9a resolve more issues 2022-03-21 17:20:05 -06:00
James Betker
9e97cd800c take the conditioning mean rather than the first element 2022-03-21 16:58:03 -06:00
James Betker
9c7598dc9a fix conditioning_free signal 2022-03-21 15:29:17 -06:00
James Betker
2a65c982ca dont double nest checkpointing 2022-03-21 15:27:51 -06:00
James Betker
723f324eda Make it even better 2022-03-21 14:50:59 -06:00
James Betker
e735d8e1fa unified_voice fixes 2022-03-21 14:44:00 -06:00
James Betker
1ad18d29a8 Flat fixes 2022-03-21 14:43:52 -06:00
James Betker
26dcf7f1a2 r2 of the flat diffusion 2022-03-21 11:40:43 -06:00
James Betker
c14fc003ed flat diffusion 2022-03-17 17:45:27 -06:00
James Betker
428911cd4d flat diffusion network 2022-03-17 10:53:56 -06:00
James Betker
bf08519d71 fixes 2022-03-17 10:53:39 -06:00
James Betker
95ea0a592f More cleaning 2022-03-16 12:05:56 -06:00
James Betker
d186414566 More spring cleaning 2022-03-16 12:04:00 -06:00
James Betker
8b376e63d9 More improvements 2022-03-16 10:16:34 -06:00
James Betker
0fc877cbc8 tts9 fix for alignment size 2022-03-15 21:43:14 -06:00
James Betker
f563a8dd41 fixes 2022-03-15 21:43:00 -06:00
James Betker
b754058018 Update wav2vec2 wrapper 2022-03-15 11:35:38 -06:00
James Betker
9c6f776980 Add univnet vocoder 2022-03-15 11:34:51 -06:00
James Betker
7929fd89de Refactor audio-style models into the audio folder 2022-03-15 11:06:25 -06:00
James Betker
f95d3d2b82 move waveglow to audio/vocoders 2022-03-15 11:03:07 -06:00
James Betker
bb03cbb9fc composable initial checkin 2022-03-15 10:35:40 -06:00
James Betker
86b0d76fb9 tts8 (incomplete, may be removed) 2022-03-15 10:35:31 -06:00
James Betker
eecbc0e678 Use wider spectrogram when asked 2022-03-15 10:35:11 -06:00
James Betker
9767260c6c tacotron stft - loosen bounds restrictions and clip 2022-03-15 10:31:26 -06:00
James Betker
f8631ad4f7 Updates to support inputting MELs into the conditioning encoder 2022-03-14 17:31:42 -06:00
James Betker
22c67ce8d3 tts9 mods 2022-03-13 10:25:55 -06:00
James Betker
8f130e2b3f add scale_shift_norm back to tts9 2022-03-12 20:42:13 -07:00
James Betker
73bfd4a86d another tts9 update 2022-03-12 15:17:06 -07:00
James Betker
0523777ff7 add efficient config to tts9 2022-03-12 15:10:35 -07:00
James Betker
1e87b934db potentially average conditioning inputs 2022-03-10 20:37:41 -07:00
James Betker
e6a95f7c11 Update tts9: Remove torchscript provisions and add mechanism to train solely on codes 2022-03-09 09:43:38 -07:00
James Betker
f56edb2122 minicoder with classifier head: spread out probability mass for 0 predictions 2022-03-08 15:51:31 -07:00
James Betker
29b2921222 move diffusion vocoder 2022-03-08 15:51:05 -07:00
James Betker
94222b0216 tts9 initial commit 2022-03-08 15:50:45 -07:00
James Betker
d1dc8dbb35 Support tts9 2022-03-05 20:14:36 -07:00
James Betker
79e5692388 Fix distributed bug 2022-03-04 11:58:53 -07:00
James Betker
f87e10ffef Make deterministic sampler work with distributed training & microbatches 2022-03-04 11:50:50 -07:00
James Betker
77c18b53b3 Cap grad booster 2022-03-04 10:40:24 -07:00
James Betker
2d1cb83c1d Add a deterministic timestep sampler, with provisions to employ it every n steps 2022-03-04 10:40:14 -07:00
James Betker
58019a2ce3 audio diffusion fid updates 2022-03-03 21:53:32 -07:00
James Betker
998c53ad4f w2v_matcher mods 2022-03-03 21:52:51 -07:00
James Betker
9029e4f20c Add a base-wrapper 2022-03-03 21:52:28 -07:00
James Betker
6873ad6660 Support functionality 2022-03-03 21:52:16 -07:00
James Betker
6af5d129ce Add experimental gradient boosting into tts7 2022-03-03 21:51:40 -07:00
James Betker
7ea84f1ac3 asdf 2022-03-03 13:43:44 -07:00
James Betker
3cd6c7f428 Get rid of unused codes in vq 2022-03-03 13:41:38 -07:00
James Betker
619da9ea28 Get rid of discretization loss 2022-03-03 13:36:25 -07:00
James Betker
beb7c8a39d asdf 2022-03-01 21:41:31 -07:00
James Betker
70fa780edb Add mechanism to export grad norms 2022-03-01 20:19:52 -07:00
James Betker
d9f8f92840 Codified fp16 2022-03-01 15:46:04 -07:00
James Betker
45ab444c04 Rework minicoder to always checkpoint 2022-03-01 14:09:18 -07:00
James Betker
db0c3340ac Implement guidance-free diffusion in eval
And a few other fixes
2022-03-01 11:49:36 -07:00
James Betker
2134f06516 Implement conditioning-free diffusion at the eval level 2022-02-27 15:11:42 -07:00
James Betker
436fe24822 Add conditioning-free guidance 2022-02-27 15:00:06 -07:00
James Betker
ac920798bb misc 2022-02-27 14:49:11 -07:00
James Betker
dbc74e96b2 w2v_matcher 2022-02-27 14:48:23 -07:00
James Betker
42879d7296 w2v_wrapper ramping dropout mode
this is an experimental feature that needs some testing
2022-02-27 14:47:51 -07:00
James Betker
c375287db9 Re-instate autocasting 2022-02-25 11:06:18 -07:00
James Betker
34ee32a90e get rid of autocasting in tts7 2022-02-24 21:53:51 -07:00
James Betker
ea500ad42a Use clustered masking in udtts7 2022-02-24 07:57:26 -07:00
James Betker
7201b4500c default text_to_sequence cleaners 2022-02-21 19:14:22 -07:00
James Betker
ba7f54c162 w2v: new inference function 2022-02-21 19:13:03 -07:00
James Betker
38802a96c8 remove timesteps from cond calculation 2022-02-21 12:32:21 -07:00
James Betker
668876799d unet_diffusion_tts7 2022-02-20 15:22:38 -07:00
James Betker
0872e17e60 unified_voice mods 2022-02-19 20:37:35 -07:00
James Betker
7b12799370 Reformat mel_text_clip for use in eval 2022-02-19 20:37:26 -07:00
James Betker
baf7b65566 Attempt to make w2v play with DDP AND checkpointing 2022-02-18 18:47:11 -07:00
James Betker
f3776f1992 reset ctc loss from "mean" to "sum" 2022-02-17 22:00:58 -07:00
James Betker
2b20da679c make spec_augment a parameter 2022-02-17 20:22:05 -07:00
James Betker
e1d71e1bd5 w2v_wrapper: get rid of ctc attention mask 2022-02-15 20:54:40 -07:00
James Betker
79e8f36d30 Convert CLIP models into new folder 2022-02-15 20:53:07 -07:00
James Betker
2bdb515068 A few mods to make wav2vec2 trainable with DDP on DLAS 2022-02-15 06:28:54 -07:00
James Betker
52b61b9f77 Update scripts and attempt to figure out how UnifiedVoice could be used to produce CTC codes 2022-02-13 20:48:06 -07:00