Q-Framework speech recognition platform
Q-Framework is a speech recognition platform developed by Code-Q. It supports local and server-based speech recognition, as well as hybrid speech recognition. The platform is especially suitable for environments where confidential information is processed.
We deliver an all-in-one solution that can be installed into the client’s own network or server, where confidential info is stored securely. This way the info related to voice input, recognition results and language model is never communicated to any 3rd parties. Basic language does not need an internet connection, as the recognition process takes place entirely within the user’s device. Q-Framework is suitable for consumer electronics, health care, industrial, home automation and automobile applications, to name a few.
With Q-Framework, speech and voice can control basically any device that is equipped with a mircophone and an internet connection. If the device’s resources are sufficient for local speech recognition, then an internet connection is not required.
For example, Q-Framework is used in Code-Q’s DialoQ line of accessability utilities, as well as Hawina Productions’ language learning games. Currently supported languages are Finnish, English, Swedish, German, and French.
A developer version of Q-Framework is available for Android, Windows, Linux and Qt.
- Independent of speaker
- Supports free dictation
- Custom wake word
- Supports sending speech commands to server in real time
- Automatic filter of ambient noise
- Supports changes to vocabulary settings while in operation
- Detects out of grammar words
- Speech dialog support
- Midspeech suggestions
- Supports midspeech switch to server-based recognition
- Adaptive and learning server solution
- Tools for creating custom language models
Code-Q Oy is a Finnish company founded in 2013. We design and manufacture technology related to speech recognition, accessibility software, as well as offering consulting services for specialized software.