Skip to content
openframeworks addon to use googles inofficial speech to text api
C C++ HTML Makefile
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
example-gstt_api_only
example-gstt_basics
example-gstt_chrome_workaround
libs
src
.gitignore
README.md

README.md

ofxGSTT

openFrameworks addon/wrapper to the Google Speech To Text API v2.

Unfortunately the first API version doesn't work anymore and API v2 is officially limited to 50 requests per day and it is still not possible to apply for a higher quota (but people reported that they were able to get up to 500 responses per day).

Anyway I can't recommend using this addon in a production environment - only for prototyping purposes.

For unlimited requests check out the workaround example via Chrome.

Feedback is really welcome!

Credits:

Originally this addon was ported from the processing library by Florian Schulz, which was based on Mike Pultz article that showed how to use the technology offered by Google without a browser.

Informations to switch from v1 to v2 were found here.

Examples:

Before you start you need to generate your own Speech API Key.

example-gstt_api_only:

not using the addon at all, only showcasing the api request with a prerecorded audio file

example-gstt_basics:

basic addon example for recording and transcripting

example-gstt_chrome_workaround

Instead of using ofxGSTT, this workaround uses the Speech to Text API within Google Chrome and sends the results to openFrameworks via WebSockets. Idea (again) ported from Florian Schulz example.

Dependencies: Google Chrome (or Chromium) and ofxLibwebsockets

Dependencies

libsndfile is included as static libs for unix(32 and 64bit). for mac you might need to brew install libsndfile and for win you can try to install the binaries provided here.

You can’t perform that action at this time.