pFad - Phone/Frame/Anonymizer/Declutterfier! Saves Data!


--- a PPN by Garber Painting Akron. With Image Size Reduction included!

URL: http://github.com/speechbrain/speechbrain/pull/2783

css" /> [DRAFT] Discrete Speech LMs by Adel-Moumen · Pull Request #2783 · speechbrain/speechbrain · GitHub
Skip to content

[DRAFT] Discrete Speech LMs#2783

Draft
Adel-Moumen wants to merge 53 commits intospeechbrain:developfrom
Adel-Moumen:gslms
Draft

[DRAFT] Discrete Speech LMs#2783
Adel-Moumen wants to merge 53 commits intospeechbrain:developfrom
Adel-Moumen:gslms

Conversation

@Adel-Moumen
Copy link
Collaborator

@Adel-Moumen Adel-Moumen commented Dec 11, 2024

What does this PR do?

Bl?

Before submitting
  • Did you read the contributor guideline?
  • Did you make sure your PR does only one thing, instead of bundling different changes together?
  • Did you make sure to update the documentation with your changes? (if necessary)
  • Did you write any new necessary tests? (not for typos and docs)
  • Did you verify new and existing tests pass locally with your changes?
  • Did you list all the breaking changes introduced by this pull request?
  • Does your code adhere to project-specific code style and conventions?

PR review

Reviewer checklist
  • Is this pull request ready for review? (if not, please submit in draft mode)
  • Check that all items from Before submitting are resolved
  • Make sure the title is self-explanatory and the description concisely explains the PR
  • Add labels and milestones (and optionally projects) to the PR so it can be classified
  • Confirm that the changes adhere to compatibility requirements (e.g., Python version, platform)
  • Review the self-review checklist to ensure the code is ready for review

Adel-Moumen and others added 30 commits November 12, 2024 16:35
@Adel-Moumen
add uLM
e0c14e9
@Adel-Moumen
modify the code structure
5435cb5
@Adel-Moumen
remove old file
1947e72
@Adel-Moumen
Merge branch 'speechbrain:develop' into gslms
b3bab5a
@Adel-Moumen
fix source
427bd67
@Adel-Moumen
prototype train.py
ef115ae
@Adel-Moumen
add data prep
9ba8ca0
@Adel-Moumen
PoC recipe
29cf041
@Adel-Moumen
add tie embds
33bef9f
@Adel-Moumen
tmp: to fix. because we hardcode the cache, then HF is not looking at…
0d3216d 
… the right cache.
@Adel-Moumen
fix the awful way of checking the cache :p
993a510
@Adel-Moumen
fix cache
8ac16ef
@Adel-Moumen
Merge branch 'speechbrain:develop' into gslms
e38c644
@Adel-Moumen
Merge branch 'gslms' of https://github.com/Adel-Moumen/speechbrain in…
79455a8 
…to gslms
@Adel-Moumen
Merge branch 'speechbrain:develop' into gslms
bb7b540
@Adel-Moumen
integration of Jarod work on preparing tokens
e7678da
@Adel-Moumen
overall logic to train an SLM
f54c4e9
@Adel-Moumen
compile
6efecbc
@Adel-Moumen
add repr/str
42523fa
@Adel-Moumen
add audio backend
aa29c45
@Adel-Moumen
packedtensor integration in train.py
ef26ce2
@Adel-Moumen
custom dataset
e9e8a7f
@Adel-Moumen
add todolist
337d2d7
@Adel-Moumen
general todos
b7ce553
@Adel-Moumen
lr annealing
5b12637
@Adel-Moumen
block_size + 1 to read max block size
b881aeb
@Adel-Moumen
seeds + ST yaml
ffa14bd
@Adel-Moumen
add fairseq twist hubert in SB
2d76951
@Adel-Moumen
small upgrades for ids and avoid + 1 for blocksize
7a720e8
@Adel-Moumen
fix hubert extraction
dca346c
Adel-Moumen and others added 23 commits January 10, 2025 11:03
@Adel-Moumen
remove useless padding as it is already done by SB dataloader with th…
bd9985b 
…e collate_fn
@Adel-Moumen
add collapse_repeated_tokens for SSL models
b2c702c
@Adel-Moumen
fix saving of empty gradscaler when using bf16
8e99c10
@Adel-Moumen
add wandb
ebd23f4
@Adel-Moumen
add rank / num_shards in yamls
a22de7b
@Adel-Moumen
catch wandb in gitignore
dabe386
@Adel-Moumen
remove useless block
b601520
@Adel-Moumen
fix pytorch warning
789c2f0
@Adel-Moumen
add vocoder in interface
fec2036
@Adel-Moumen
initializer directly in the WandbLogger class
1154735
@Adel-Moumen
move backbone aside + wandb log better init
c987e92
@Adel-Moumen
backbone aside
56f6eb1
@Adel-Moumen
naming + removing data prep as it should be done here
b98b323
@Adel-Moumen
finally rename ssl to speechlm lmao
07e8de1
@Adel-Moumen
update wandb
b6aceb4
@Adel-Moumen
fix empty gradscaler + initial step
9395af8
@Adel-Moumen
packedtensor + general improvements
5cfe8e7
@Adel-Moumen
dummy stuff will be removed latter
3d21a39
@Adel-Moumen
Merge branch 'develop' into gslms
44360fa
@Adel-Moumen
Merge branch 'speechbrain:develop' into gslms
1903805
@Adel-Moumen
update codebase
4204fd9
@Adel-Moumen
sq codec
e73df17
@Adel-Moumen
Merge branch 'gslms' of https://github.com/Adel-Moumen/speechbrain in…
42ee420 
…to gslms
@TParcollet TParcollet added this to the v1.1.0 milestone Oct 9, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

v1.1.0

Development

Successfully merging this pull request may close these issues.

2 participants

@Adel-Moumen @TParcollet

Comments
pFad - Phonifier reborn

Pfad - The Proxy pFad © 2024 Your Company Name. All rights reserved.




Check this box to remove all script contents from the fetched content.



Check this box to remove all images from the fetched content.


Check this box to remove all CSS styles from the fetched content.


Check this box to keep images inefficiently compressed and original size.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy