Cloud Speech Recognition
The smart mirror uses [Sonus](https://github.com/evancohen/sonus) for speech recognition. The following steps will get you everything you need to make this work:
When prompted to create a new service account select "Project Owner"
After setting up your project you will also need an API key for
- youtube (if you intend to use this function on your mirror)
enter part of the name of each api in turn, and select it, and click enable repeat for the other apis listed after completing the enabling for all the listed apis
- from the 3 bar menu on the left, select APIs and Services, then Credentials
- in the Create credentials dropdown, select "API key",
- save the value that is shown on the message window that pops up
The API key should be a long string, something like this 'AIzaSyBGWRB2oC1P_xYteMMsdTRV3Kv3aEBPQg'
Note: This is no longer required, the mirror comes with a pre-trained universal model! If you're having trouble with the universal model you can follow these steps
Training your own keyword will drastically increase the accuracy of keyword detection. This will be most accurate if you do this on your Pi using the microphone that you'll be using to trigger the mirror.
From the smart mirror directory run:
you can train your model here: [https://github.com/seasalt-ai/snowboy]https://github.com/seasalt-ai/snowboy)
Once trained, download the model and save it to the root of the smart-mirror directory.