The DeepSpeech project started by Mozilla has updated its GitHub page with the message "This project is now discontinued", and a change in the project status to archived. 

It's a sad end to an interesting project. DeepSpeech was an open source embedded (offline, on-device) speech-to-text engine which could run in real time on devices ranging from a Raspberry Pi 4 to high power GPU servers. DeepSpeech used a model trained by machine learning techniques based on Baidu's Deep Speech research paper. The project used Google's TensorFlow to make the implementation easier.

The plan was that DeepSpeech would have a transcription word error rate of lower than 10%, and it achieved that aim for its pretrained English-language model with a word error rate of around 7.5%. 

Mozilla launched DeepSpeech in 2017 as an incubation project by the machine learning team at Mozilla Research. The aim was to create an open source automatic speech recognition model. Work was going well until 2020, when Mozilla announced a layoff of approximately 250 employees and a big restructuring of the company. At the time Reuben Morais, Senior Research Engineer working on the Machine Learning team, said that the team was working to find out if the project would have a new home in the restructured Mozilla, and what changes would be necessary for a successful transition.

At the time the team was gearing up for their first stable release, version 1.0., Morais described DeepSpeech as being an easy-to-use, open source speech-to-text solution that can be easily integrated in many platforms, programming languages and types of applications, saying that no other open source solution came close to the accuracy, maturity and ease of use of Mozilla's  tools.

Following that move, in 2021 Mozilla changed from active development to an advisory role including launching a grant program to fund initiatives showing applications for DeepSpeech.

Sadly, the developers have now closed down DeepSpeech. 

More Information

DeepSpeech On GitHub

Related Articles

Introducing DeepSpeech

Mozilla Updates Voice Recognition Project

Mozilla DeepSpeech Gets Smaller

Mozilla Layoffs - The Fallout

 Whisper Open Source Speech Recognition You Can Use

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info