Flutter App for WMS
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Nicolas Lenz e0e827217c Update Readme 1 day ago
android Fix activation voice prompt 1 day ago
lib Help message on voice activation 1 day ago
.gitignore Upload 1 month ago
.metadata Upload 1 month ago
LICENSE Add Apache License file 4 days ago
README.md Update Readme 1 day ago
pubspec.lock Normalize locations, use DEFAULT audio input for hotword 4 days ago
pubspec.yaml Bump version to 1.1.0 1 day ago



The voice-controlled warehouse management. To be used with wms-server. Implemented using Flutter.

Requirements and Building

Currently only works on Android, requires Flutter and a working Android SDK. Compilation can be done with flutter build apk on Linux in the cloned repo. The APK is then found in build/app/outputs/apk/release/app-release.apk.

Prebuilt APKs can also be directly downloaded in the releases sections.

The Google app (yes, the search app) must be installed on the target device as it contains the speech recognition APIs.


Activate listening (after setting up credentials, of course) by pressing the microphone button and saying “Hey, Pico”. You can then say a command after the “pling” tone. Currently available commands are: “pick”, “put away”, “help” and “repeat”. The rest of the following dialogue should be self-explainatory.

Say “cancel” at any point while recognition is on (it’s after a “pling”) to cancel the operation. You can start over then. Also, while saying something, you can add an “again” before the closing “pling” and the app will let you say it again.

When answering most dialogue questions, adding words is possible without problems. So “I would really like to pick some containers” will work just as well as a plain “pick” (maybe even better as recognition works better with longer sentences with context).

Container numbers can only be numbers, location names can be combinations of numbers and letters. Case and special characters are ignored.


Uses Porcupine as a library for hotword detection.