85 Commits

Author SHA1 Message Date
autumn-2-net 76afe57e47 New generative model algorithm: Rectified Flow (#184)
Implements Rectified Flow in DiffSinger.

ref:
- https://github.com/gnobitab/RectifiedFlow
- https://github.com/yxlllc/ReFlow-VAE-SVC

---------

Co-authored-by: yqzhishen <yangqian_1015@icloud.com>
2024-04-17 22:57:54 +08:00
yqzhishen 73c54c91ad Change default mel_base to e
This completes stage 1 of log base migration plan
2024-04-17 21:54:02 +08:00
yqzhishen e17c428d42 Finish docs for v2.3.0 release (#177)
* Complete configuration schemas for tension and voicing

* Add docs for choosing variance parameters
2024-03-11 00:52:22 +08:00
yqzhishen c16095bc55 Drop support for some old features and behaviors (#172)
* Drop support for discrete F0 embedding (reserved in ONNX exporter)

* Drop support for `interp_uv` configuration key

* Drop support for `train_set_name` and `valid_set_name` configuration keys

* Drop support for linear domain of random time stretching augmentation

* Drop support for `num_pad_tokens` configuration key

* Drop support for code backup before training

* Drop support for `ffn_padding` configuration key

* Drop support for random seeding

* Add placeholder to load old checkpoint

* Remove duplicate txt_embed layer (resuming may raise errors)

* Remove migration script and error message for transcriptions.txt

* Remove seed from batch shuffling

* Use direct access on some hparam keys

* Fix duplicate keys in YAML

* Rename `pndm_speedup` to `diff_speedup`
2024-02-25 20:57:02 +08:00
yqzhishen c2b8e75b6d Sync image version 2023-12-08 00:08:40 +08:00
yqzhishen a1edc18a8d Update architecture graph with melody encoder 2023-12-07 23:55:58 +08:00
yqzhishen 246c286ee0 Enable shallow diffusion and AMP by default 2023-12-07 21:23:22 +08:00
yqzhishen a22ace1f8e Add notice for TensorBoard issue with DDP 2023-12-04 23:35:02 +08:00
yqzhishen 18f4cbdb21 Improve error message and add instructions for NSF-HiFiGAN exporting 2023-12-02 20:02:38 +08:00
yqzhishen 07908243fa Update f0_max to 1100 and sync with docs 2023-12-02 18:44:12 +08:00
yqzhishen baa48221d0 Adjust documentation about materials and pre-trained models 2023-12-01 20:51:14 +08:00
yqzhishen 575ff7ba4c Add an introduction of mutual influence between variance modules 2023-12-01 14:07:56 +08:00
yqzhishen 006dad395b Finish documentation for v2.2.0 release (#156)
* Finish ConfigurationSchemas.md for v2.2.0 release

* Add detailed instructions for shallow diffusion

* Update link to DiffScope
2023-11-22 10:35:38 +08:00
UtaUtaUtau 931df279e0 Add WORLD pitch estimators and F0 range as hyperparameters (#149)
* Added WORLD pitch estimators

i also removed hardcoded F0 ranges because what the heck is that 800 Hz max pitch in parselmouth that is way too low

* Update README.md

okay maybe don't add ur own flare in the readme if u actually want to create a pull req

* Apply dtype change

i saw it in the parselmouth thing might as well put it in to make sure

* Update pw.py

oops

* fix dtype mismatch

for some reason pyworld only likes float64?

* add f0 range as a hyperparameter

why isn't it a hyperparameter in the first place

* move pad_frames to pw

i think world is p accurate with the frames stuff but it's just to ensure

* change padding algorithm

it's just to be similar to the parselmouth one.. it makes sense to not center the F0 after all

* remove duplicate line

yeah

* Remove DIO, change default range, add docs

* Add notice for F0 range
2023-11-21 00:04:24 +08:00
Dachun Sun b42dcbf72e Update BestPractices.md 2023-11-04 10:21:06 -05:00
Dachun Sun ae2a50fafc Update documentation formatting (#153)
* Update doc formatting

* HTML table

* Fix links in HTML

* Fix links

* Fix links
2023-10-29 12:22:35 +08:00
Dachun Sun 36ac5f583b Multi-node batched validation and improvement on strategy selection (#148)
* More metadata when binarize

* Picklable AttrDict

* Support other notion of 'sizes'

* Better strategy getter, Ds TB logger, and eval batch sampler.

* Add title to plots and specify figsize

* Unify batch sampler and bug fixes

* Batch and multi-device validation

* Fix imports

* Fix deadlock under multigpu and resume from ckpt

* Remove unnecessary if in base_dataset

* Move build loss to finish init and ft to build model

* Prevent repeated valid item

* Remove optimizer_idx to support lightning 2.1

* Warning message fix

* Fix val error when aux is off

* Rename fields in metadata and add doc

* Adjust respective duration logging for each speaker

* val persisent worker, module list ordering, update doc

---------

Co-authored-by: yqzhishen <yangqian_1015@icloud.com>
2023-10-29 00:37:14 +08:00
yqzhishen 4957eb23db Update RMVPE pretrained mode l download link 2023-10-08 23:05:51 +08:00
yqzhishen 662136d237 Update RMVPE checkpoint download link 2023-08-26 13:43:56 +08:00
yqzhishen 57087ca6e3 Add docs for optimizers and LR schedulers 2023-08-24 21:44:41 +08:00
yqzhishen 4d1d332a97 Add docs for DS binarization 2023-08-24 21:41:21 +08:00
yqzhishen 005f666d79 Add docs for fine-tuning and parameter freezing 2023-08-24 02:02:41 +08:00
yqzhishen ffebdcdb8a Add docs for pitch extractors 2023-08-21 02:59:57 +08:00
yqzhishen 1004625f23 Add -mixed 2023-08-14 16:56:04 +08:00
yqzhishen 2e1dd5a8af Add TensorBoard usage 2023-08-14 02:22:51 +08:00
yqzhishen 8608009176 Finish overviews of acoustic and variance model 2023-08-14 02:22:39 +08:00
yqzhishen 5861211b86 Finish introduction for materials, add tips 2023-08-13 02:32:30 +08:00
yqzhishen 6fc85f8519 Add missing command 2023-08-11 17:28:49 +08:00
autumn-2-net 1031f78714 Support freezing model parameters during training (#120)
* fix some

* support freeze model params

* support freeze model params

* support freeze model params

* Rename parameter and add docs

---------

Co-authored-by: autumn <2>
Co-authored-by: yqzhishen <yangqian_1015@icloud.com>
2023-08-04 19:41:15 +08:00
yqzhishen f74046db31 Support custom pitch extractor (#118)
* Support custom pitch extractor

* Add doc for `pe` and `pe_ckpt`
2023-07-19 23:57:38 +08:00
autumn-2-net 7847af2d11 support finetuning from pretrained checkpoints (#108)
* support pre_train model add doc

* Update docs for finetuning

---------

Co-authored-by: autumn <2>
Co-authored-by: yqzhishen <yangqian_1015@icloud.com>
2023-07-17 21:24:22 +08:00
yqzhishen bae626f174 Add configuration templates 2023-07-16 18:40:35 +08:00
yqzhishen cdcfb880a6 Data distributed -> distributed data 2023-07-14 17:14:40 +08:00
yqzhishen d6a5addc3d Adjust docs 2023-07-13 00:54:14 +08:00
yqzhishen 608858d17a Finish documentation for version 2.0.0 (#106)
* Refactor README

* Delete old docs

* Adjust diagram

* Update image in README.md

* Add images

* Update references

* Update README.md

* Finish GettingStarted.md

* Remove redundant paragraph

* Remove docs

* Remove images

* Remove unused images

* Add link to Releases

* Add link to discord server

* Add explanations to the configuration system

* Add custom dictionary guidelines

* Server -> server

* Add performance tuning guidelines
2023-07-13 00:38:01 +08:00
yqzhishen 21d160e93f Update diagram 2023-07-09 20:35:31 +08:00
yqzhishen af06e30d11 Update architecture graph 2023-07-04 01:34:48 +08:00
yqzhishen 8bd93d15a9 Add draw.io files of architecture graphs 2023-06-30 02:11:05 +08:00
yqzhishen b2f9aaffef Change repeat_bins to total_repeat_bins 2023-06-24 23:11:53 +08:00
yqzhishen 65b4be379b Modify param names and docs for ds_id and spk_id 2023-06-24 02:36:49 +08:00
yqzhishen 224fd33f39 Support custom spk_id arrangements 2023-06-24 02:20:51 +08:00
llc1995@sina.com cff5da2f29 Add DPM-Solver++ and UniPC sampling method 2023-06-17 21:23:18 +08:00
llc1995@sina.com 76602951d3 add DDIM sampling method 2023-06-17 01:50:29 +08:00
yqzhishen c4dbe37925 Add missing diff_accelerator 2023-06-03 23:29:33 +08:00
yqzhishen a4d5221168 Complete configuration schemas for variance task 2023-06-03 21:27:24 +08:00
yqzhishen 6f44a321df preserved -> reserved 2023-05-17 19:31:41 +08:00
yqzhishen 99f10793aa Change default num_pad_tokens to 1 2023-05-17 19:27:33 +08:00
yqzhishen 2b080d04ed Adjust descriptions 2023-05-13 21:22:37 +08:00
yqzhishen 34fe5399ff Augmentations can now be disabled with enabled config keys 2023-05-13 21:17:36 +08:00
yqzhishen a89d3ee6f6 Add configuration references documentation (#86)
* Add ConfigurationSchemas.md (#77)

* Update ConfigurationSchemas.md

Explain configuration parameters in detail

* Update ConfigurationSchemas.md

* Update ConfigurationSchemas.md

* Update ConfigurationSchemas.md

Add DDP configuration details

* Update ConfigurationSchemas.md

Update Neural Network configs ( 3 items remain TBD)

* Update ConfigurationSchemas.md

Add others

* Refactor ConfigurationSchemas.md

* Finish ConfigurationSchemas.md

* Finish and format ConfigurationSchemas.md

* Reduce long explanations and fix mistakes

* Fix math expressions

* Fix math expressions

* used by -> visibility

* Adjust descriptions

---------

Co-authored-by: Weiyang <frank.wang04@gmail.com>
2023-05-13 21:00:39 +08:00