Wake Word Creation Tool
Wake Word allows to voice-enable the activation of your products and services. Use our custom creation tool to make your own embedded trigger word.
What is a Wake Word ?
Wake Word is a passive listening system that has only one objective, to detect the keyword assigned to it. The main interest of this tool is to save the resources used by activating the actual listening of the user only when the wake word has been recognized.
For instance, “Hey Siri” or “Ok Google” are reknown hot words. What don’t you make one of your own ?
Totally embedded, therefore private by design, our technology does not record any voice data during listening.
Words must be chosen according to different criteria to be effective and avoid false positives (bad detections). First of all, choose words with several syllables to make it more robust and less prone to errors.
Choose a keyword that identifies with your brand to ensure your image and notoriety. The chosen keyword(s) must be as unique as possible in the spoken language to avoid accidental triggering.
Our Must-have features
Our solution is able to work with 36 different languages, the most commonly spoken (English, Mandarin, Spanish….) in order to scale your solution worldwide.
Wake Words are already popular with big brands such as GAFAs. We offer you to produce your own trigger word with your brand name in it.
Wake Word is a totally embedded technology that does not rely on internet connection or 3rd parties. It is therefore Private by Design and GDPR-Compliant
Very low CPU usage
Our technology occupies a minimal CPU load to operate. This capacity comes from its compact design, specialization, and the technical performance of the tools used to create it.
Some of our clients use our technology
Wake Word is a commonly used technology in the word of voice interfaces and assistants. Some of our clients, in order to achieve their projects and innovations, has developped interesting features thanks to our embedded wake words.
OS-dependent API binding and packaging
- Android: WUW (within VDK) will be served with a Java-API binding compiled into an Android archive (AAR)
- Win/Linux: WUW (within VDK) will come with C++ API binding
- Apple iOS: WUW (within VDK) will come with Swift binding
Standard ports and Tools
- iOS (version 7.0 and up): arm64 and x86_64
- Android (version 5.0 and up): armv7 (32Bit), arm64 and x86_64
- Linux: armv7 (32Bit), arm64 and x86_64
- Windows: x86_64
Components and relative data size per language
- Accoustics models, per language
- Gen 4 compact: 900kB
- Gen 5: approx. 4MB
- Gen 6: approx. 6MB
- GLIC – mono-lingual – General purpose transcriptions: 300-7300kB
- GLC – multi-lingual – Music collection compilation: 700-3000kB
Our Wake-up Word is an embedded technology that is made to be integrated into devices. To do so, these products need to meet specific criteria to handle the WUW and make it work properly to perform your use case.
Frequently Asked Questions on Wake Word
A few things to know…
This solution can be tricky since it is a complex technology. We cover some of the recurrent topics about it in order to give you insights.
How to make a good performing wake word for my use case?
You need to find words that are simple (easy to remember), uniform (doesn’t change) and unique (not causing false-positive)
Does the wake word listen and record everything said?
Our solution work by passively listening (not recording) their suroundings until they detect the keyphrase/word.
Are there any risks of false-positive while using trigger words?
Any risks can be lowered at its maximum. Choice of wording will determine its robustness and the probability of a wrong activation