4

I'm making a project that requires the user to control the computer through speech. I've seen a couple of libraries, for python, that can do this but most of the ones that I have seen are windows only. There is a library called dragonfly which seems to be very popular but I havn't seen anything telling me that it's windows only.

Are there any Raspberry Pi compatible libraries that you know of?

Jivings
  • 22,538
  • 11
  • 90
  • 139
one2three
  • 1
  • 1
  • 4
  • There is a raspberry pi project called Jasper dedicated to doing speech recognition on the Pi, but I have yet to try it out http://jasperproject.github.io/ – rob Nov 10 '14 at 10:16
  • Yeah it looks good but it seems to be an os. For my project I need to still have the screen under my control. My project is a smartphone type device but the main controls are voice/typing. The user would request for things to be shown on the screen. It's kind of lie google glass but it'll have features that I think should be on google glass(and it'll save me 1000 pounds as Explorer edition is the only available google glass in the UK). – one2three Nov 10 '14 at 20:27
  • related: http://stackoverflow.com/questions/28533764/python-speech-recognition-for-raspberry-pi-2 – Ghanima Jan 05 '16 at 13:28

4 Answers4

2

Yes there is Pocketsphinx to convert speech to text, that has python APIs

For installation do:

apt-cache search pocketsphinx

and install all of them.

Or I have a shorcut to do this: :D

apt-get install $(apt-cache search pocketsphinx | \
awk '{print $1}' | tr '[:space:]' ' ')

enjoy.. :)

Bex
  • 2,929
  • 3
  • 25
  • 34
RahulAN
  • 149
  • 7
1

Apologies for the late answer, but hopefully someone will find this useful.

I am using this python package on my pi, and its fantastic!

Uses the google recognition engine (the one used in chrome and "okay google") which is arguably the best for accuracy, as well as supporting free form dictation which depending on your needs, could be a real benefit (would allow for example googling a non pre-determined phrase)

Only drawback is that it requires an internet connection, and the API key is only for personal use and development, so if you are planning on launching a commercial product, this will not be viable for you.

Hope this helps.

Aphire
  • 217
  • 4
  • 15
0

I'm using pocketsphinx and highly recommend it for offline speech recognition. Here you have the instructions to install it: http://cmusphinx.sourceforge.net/wiki/raspberrypi

Also have a look at this question, just in case you run into the same problems I did to make it work.

jotadepicas
  • 394
  • 4
  • 15
-4

Is this what you want? Google is your friend.

leo60228
  • 121
  • 2