C library for speech recognition using the Bing Voice Recognition API
C C++ Makefile
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
examples
include/bingsprec
src
LICENSE
README
makefile

README

libbingsprec
----------------
If you want to leverage a speech-to-text API on devices 
like the Raspberry Pi, then this library is for you. While there 
are other similar libraries, all of them use Google's STT API 
which has a relatively small rate limit (50 queries/day). Bing's
public API, on the other hand, has a limit of 5,000 queries/month 
on the free tier, with the option to pay for more requests if 
needed.

Feel free to make improvements and submit pull requests!

How To Build:
-------------------

1. Register for Bing Voice Recognition API (Beta)
	http://datamarket.azure.com/dataset/bing/speechrecognition

2. Edit include/bingsprec/constants.h
	- Replace BINGSPREC_CLIENT_ID with your client id
	- Replace BINGSPREC_CLIENT_SECRET with your client secret
	- (In most cases) leave the other values alone

3. Install the required libraries
	- libcurl: sudo apt-get install libcurl4-openssl-dev
	- uuid: sudo apt-get install uuid-dev

4. Build bingsprec
	- sudo make install

5. Build the example
	- sudo make simple

License:
-------------------
GNU LGPL-licensed. See LICENSE file.

Credits:
-------------------
Inspiration from H2CO3's libsprec, which uses Google STT API instead of Bing's API.
https://github.com/H2CO3/libsprec