I am looking for a solution to enable offline voice control for my smart home devices without the use of a wake word. The system should allow me to control devices by simple voice commands, such as “turn on the light,” “turn off the air conditioner,” or “open/close the curtains,” without needing a specific wake word.
Requirements:
No Wake Word: I want the system to operate without the need for a wake word (e.g., “Hey Siri,” “Alexa,” etc.). Voice commands should be recognized without needing any specific activation phrase.
Offline Functionality: The system should work without an internet connection. However, if an offline solution is not feasible, an online solution would be acceptable, but it must still not require a wake word.
Microphone-only Setup: I only want to use a microphone device for voice input, with no speakers or any form of audible feedback.
Smart Home Gateway (Central hub for controlling devices)
Various smart home devices (lights, air conditioner, curtains, etc.)
Previous Attempts:
I have tried using HAOS combined with the RHASSPY plugin, but encountered several issues:
ReSpeaker 4-Mic Array is not recognized correctly by HAOS and the RHASSPY plugin. There seems to be a driver issue, as I couldn’t install it properly.
When attempting to use the recording command in SSH, it fails due to missing commands, preventing the microphone from functioning as expected.
Request:
Could anyone provide a new solution or technical approach to achieve this setup, either using Home Assistant or other compatible systems? I would greatly appreciate any suggestions or guidance.
For the audio part without wake word, i think of two solutions.
use some streaming tts (kaldi, if you want to run on cpu) which will transcript everything, and when you detect intent in the transcription, do the command.
hongge_li, Rhasspy and Home Assistant both are designed to work without an internet connection. However I wonder if this will be as good as you expect.
Having the wakeword and audio feedback helps to make the experience feel like talking to a human assistant. I think that otherwise it could feel pretty spooky having things happen when you didn’t intend them.
I find the audible feedback invaluable to know whether Rhasspy has woken up and is listening or whether it understood the command. I find sometimes I have to say the wakeword a couple of times before I can give the command - and without feedback, I imagine it could be very frustrating when rhasspy doesn’t respond as expected.
Also having HA respond to parts of your conversation it is overhearing - but not directed at Rhasspy/HA - may seem strange or even rude. Imagine for example asking your partner “Would you like the light on ?” and having it happen even though your partner replied “no”.
Consider also that Rhasspy’s speech recognition is far from perfect, resulting in false positives where it thinks it recognised something that you didn’t actually say. And even with a wakeword my automation to turn the TV on also turns off the Rhasspy device in the living room to avoid false positives from the TV dialogue.
