Hi,
Where can I find a full guide on how to use Deep Speech.
Can I use it within a mobile app?
Any guidance, please ?
I am working on my graduation project. It is a mobile app that would classify a speech (an arabic speech to be precise) as healthy or pathological. The classification method would be as described here
The documentation on github has more on using the tf-lite model instead of the full size version. I’m sure you’re looking for more than that but the DeepSpeech.py uses an if statement to determine if tf lite or not.
There is also a deepspeech-server in python if inference was to be done online vs offline.
Thank you for writing back.
Really I am not an expert, whenever I check the GitHub repo for Deep Speech I feel lost.
hi @tuttlebr,
I think that my tuto could be a good start, even if some changes occured with newer deepspeech versions.
And, of course, tf lite could be better for any mobile need.
Oh that’d be cool. Where can one find your tutorial?
But as i told you, you ll have minor changes to operate, fo success
Guys, you are amazing.
@elpimous_robot I’ll check your tutorial, and post any advancement.
Thank you
… and welcome here !!!
You’ll see, Mozilla and specially Deepspeech team, always help and work hard to produce the best STT (and free !!)
If I understand this, you are training a model. Are there existing pretrained models available? How would I use one on ubuntu? I want to use my microphone and do STT to run commands. Cheers! M
actually, there is (only for English though). You can find instructions on how to use it on DeepSpeech GitHub page at https://github.com/mozilla/DeepSpeech/blob/master/USING.rst#using-a-pre-trained-model