Open JRMeyer opened 3 years ago
@JRMeyer Agreed very much with you that it gave a completely wrong expectation to those who were looking for code update !!!!
Sorry, it is not a great idea. Our company created a huge codebase based on this repo and we are actively looking to contribute it back to the Mozilla community. We still see DeepSpeech as a great project!
@bobkleiner -- archiving this repo (in the Github sense) shouldn't affect your ability to release your own codebase... where do you see the issue?
Archiving allows this repo to persist, but it also lets contributors know that PRs and open issues aren't being reviewed.
Why do we need another repo? We are ready to contribute to this.
Because nobody is around to review and merge your contributions to this repo.
@reuben Yup, the hard fact is the core team of DeepSpeech has been laid off. Hence, no one is going to support it anymore.
Because nobody is around to review and merge your contributions to this repo.
Mozilla is looking for a new team to support DeepSpeech, so things could change soon.
Because nobody is around to review and merge your contributions to this repo.
Mozilla is looking for a new team to support DeepSpeech, so things could change soon.
Could you please elaborate from where you have that information ?
Does it possible for Mozilla to facilitate the recruitment of new maintainer volunteers? (could be from inside and outside Mozilla). I guess I see some potentials in above comments.
Does it possible for Mozilla to facilitate the recruitment of new maintainer volunteers? (could be from inside and outside Mozilla). I guess I see some potentials in above comments.
This is work we tried to push and complete last year until the early days of 2021, however nobody ever stepped up on the call top volunteer and actions we put in place. Moving CI from TaskCluster to GitHub Actions move was also a big milestone we achieved to help people take ownership.
It looks like that ship has sailed, and now Mozilla has nobody working anymore on the project full time, and while it's not impossible for people to review PRs, it makes it much much more complicated.
While looking to get high-quality STT working at home, on an NVidia Jetson board, I saw mention of the possible deprecation of Deepspeech. The following 'fork' of Deepspeech (as of v0.93) was introduced, supposedly developed by former members of Mozilla's Deepspeech team:
In 2016 while at Mozilla the founders of Coqui noticed that speech technology was siloed in large corporations....
Code is at https://github.com/coqui-ai/STT
May you consider to transfer maintenance to @abandonware community best effort
what is happening with the Repo? can't even install deepspeech library.
what is happening with the Repo? can't even install deepspeech library.
no wheel at python 3.10......sad
So is DeepSpeech dead, nor we have any alternative?
Coqui STT is shutted down too.
So we don't have any FOSS speech-to-text software?
So is DeepSpeech dead, nor we have any alternative?
Coqui STT is shutted down too.
So we don't have any FOSS speech-to-text software?
👋 Hello!
It seems that DeepSpeech is no longer actively maintained, but there's no explicit message to users. Current and new users continue to send PRs and open issues with the expectation that the project is still active.
Github actually offers a nice, built-in solution for projects which are no longer actively maintained:
archiving
. As the github docs explain:This way, the code is still available to everyone, but sets expectations for users that there will be no updates.
It might also be helpful to include a link in the README to the actively maintained fork of DeepSpeech.