anonym-tsk / mmm-hotword Goto Github PK

MMM-Hotword

MMM-Hotword is a hotword detector using snowboy. You can use this module to wake another voice assistant or to give a command to other module.

Screenshot

UPDATES

2.1.0 (2020-01-13)

• Installer being updated. (Now it will check your dependencies and sound environment)
• New recipe for MMM-AssistantMk2 (v3) being added.

How to update (From previous version 1.X)

• You need to remove old MMM-Hotword directory then re-install from scratch again.
• Configuration should be re-written. It is not compatible with old version.

Installation

Read Wiki:install

(OPTIONAL) Get your personal model (.pmdl)

Instead using universal model, You can make your own personal model. (ex. Hey,Dude, volume up, ...)

Read wiki:trainer

• Personal model is dedicated to each specific vocal pattern. You cannot share your model with others.
• Personal model should be created on target device and mic. It might not work on other device.

Or.. For universal model (.umdl)

Read wiki:models

UPDATE HISTORY

2.0.3 (2020-01-08)

• Create auto-installer via npm install for RPI/Linux (manual installation is always available)
• Create post install script for check Mic configuration

2.0.2 (2019-12-04)

• Fix: issue of HOTWORD_RESUME notification not working
• Fix: some bugs of omitted configuration field.
• Add: some more front-end log to check command working

2.0.1 (2019-05-20)

• Fix: issue of afterRecording couldn't stop
• Add: afterRecordLimit to stop recording by force.

2.0.0 (2019-05-19)

• Whole new build-up
• Some annoying dependencies are removed.
• Installer is provided. (installer/install.sh)
• Personal model trainer is provided. (trainer/trainer.sh)
• Continuous recording after hotword detection is supported (Now you can say like "Computer, volume up" without pausing between Computer and volume up)
  • This feature could be used with MMM-AssistantMk2 ver3.x(Not yet released, but will arrive soon)
  • Or you can use this feature with other AI or Speech-To-Text program.
• Simple standalone commands could be available. (Without any Assistant or Speech-To-Text, you can make own voice commands with this module standalone.). Commands could be combined sequence (You can make "volume-louder" with voice models "volume" and "louder")
• More universal models are added. (computer, subex, hey extreme and more.)
• Hotword detected could be displayed on screen of MM.

1.1.0 (2018-11-4)

• notification configurable. (You don't need MMM-NotificationTrigger any more for using with MMM-AssistantMk2(^2.0.0))
• But if you want more complex action chains, you can still use MMM-NotificationTrigger also.

Last Tested; (2019-05-19)

• MagicMirror : v2.7.1
• Tested Environment :
  • Raspbian Stretch (Raspbian 3B+) / node v8.16.0 / npm v6.4.1
  • TinkerOS (TinkerBoard)
  • Ubuntu 18.04 (NVIDIA Jetson Nano)
  • OSX 10.14.4 (Apple MacBookPro) / node v11.12.0 / npm v6.7.0