Voice Recognition Attribute

From Home Remote
Jump to: navigation, search


Voice Recognition[edit]

For voice recognition to work the iOS app must be open. This is an iOS limitation. Home Remote uses the OpenEars library for iOS which supports the English language only. Other languages might work, but are not supported. Commands are activated by using the vocal prefix, by default Home Remote, then saying the action name. You can change the vocal prefix in Settings. When activating this setting for the first time you will need to grant Home Remote access to the microphone.


ScreenVoiceRecognition.jpg

Voice Recognition[edit]

Voice recognition is activated on a per-action basis. Home Remote will not activate the microphone unless at least one action has voice recognition turned on. The app will only activate the microphone whilst Home Remote is in the foreground.


Custom Vocal Name[edit]

Normally the voice activation trigger for an action is the action name given to the action. If you would like a different name, for easier activation, then type it in this field. You will be able to activate the action with the original name or the custom vocal name.


Instructions[edit]

These instructions will display the vocal prefix and action name, or custom vocal name if there is one.


Amazon Echo[edit]

So far there are 3 ways to have Amazon Echo support, each with their own pros and cons:

Home Remote - Activator can interface with the IFTTT Maker Channel as it has a webs server built in. This means you can port forward your external IP to the Home Remote server and then use that to call your Indigo actions. https://youtu.be/IjF7CANJNQI

Pros: It's quick to call: "Alexa, trigger my study lights on" Cons: It's rate limited, and sometimes there is a delay.

The second method to call indigo was to write a custom skill, but as Jay noted, there is no way to send a configuration to a skill, meaning no server address or custom utterances. I will be releasing my skill on GitHub for people, along with setup instructions. You can see a video here: https://youtu.be/RwkkHBU8MZg

Pros: It's accurate, and feels better. You won't need to create an IFTTT trigger for each actions. It's immediate. Cons: You will need to set it up yourself, although this is only 20 minutes of following instructions.

The third way is using something like the Hue or WeMo scripts that will fake a device and then tie actions to it. I use the fauxmo Python script You can see a video here: https://youtu.be/cSPBfL62XPU

Pros: It's accurate ,fast and feels better. You won't need to create an IFTTT trigger for each actions. It's immediate, with no external network calls Cons: You will need to set it up yourself. You are limited to 16 devices, each with an on and off URL call.