A repository for demos illustrating features of the Web Speech API. See Web_Speech_API for more details.

Run recognition demo live

This demo works pretty nicely — you'll need a phone running Firefox OS 2.5+, and Firefox Nightly/Developer edition, or Chrome/Chrome Mobile. Use WebIDE to install the app on your phone and give it a go. You tap the screen then say a colour — the grammar string contains a large number of HTML keywords to choose from, although we've removed most of the multiple word colours to remove ambiguity. We did keep goldenrod, cos, well.

Note that permissions are required to use speech recognition on Firefox OS.

Note also that to get this code running successfully on Chrome/Chrome mobile, you'll need to run the code through a web server (localhost will work.)

##Phrase matcher demo The same comments as above apply to this demo too — this is another demo that relies on speech recognition, written for a research team at the University of Nebraska at Kearney.

Run phrase matcher demo live

Run synthesis demo live

This demo works even better — speech synthesis needs no permissions to work on Desktop Firefox or Firefox OS. In this demo you can choose a voice from the available synthesis voices on your device. You can then enter a phrase you want to hear spoken in the text box, and press return to hear it.

1. The SpeechRecognition.continuous property is not currently implemented in Firefox. This work is ongoing.
2. The onnomatch handler does not yet work properly in Firefox — the speech recognition system always returns a positive match, and then guesses at what item in the grammar it found. This will be worked on at some point soon hopefully.