@carteblanche [https://groups.google.com/a/kitt.ai/forum/m/#!topic/snowboy-discussion/UIX2eK4ScjI](Recompile for custom wake word)
You need to create your own voice model and recompile the code. Check the above link. It is pretty straight foward though I found that I get lots of false hits but that’s because my model number doesn’t have a lot of samples.

New update is sick. Really like the turn display on/off feature.

@Mykle1 Sorry to hear it’s not working.
When you reboot try it without running the start-pulseaudio-x11 command. If that doesn’t work, I’m lost.
Iirc, when I set mine up, I paired the speaker, then did the right click on the speaker icon to change it to the bluetooth speaker.

Just did the survey.
Good luck with your thesis.

@Mykle1 try the command line instead of the gui to pair your speaker.
Run bluetoothctl in terminal.
Then type:
power on
Default-agent
Agent on
Scan on
then see if bluetoothct sees the speaker.

@Mykle1 Can you try the right click on the speaker icon now to see if it works?

@Mykle1 Getting close.
Run pulseaudio --start then try the speaker thing again.

@Mykle1
Hey,
I was able to get bluetooth audio to work on Raspbian stretch. I tried it last weekend and ran into some issues but I was able to solve(with some help from the raspberry pi forum) it this morning. I’ve had a stable connection for about 3 hours.

Yes, my pi runs MM with about 10 modules and the amazon alexa sdk.
I documented the process here.
[link text]https://www.raspberrypi.org/forums/viewtopic.php?f=66&t=198501(link url)

@Mykle1 I agree with you on that.
There’s no reason to upgrade to stretch unless you really need to. In my case, I had to upgrade to stretch because the latest alexa sdk doesn’t support jessie. When I say upgrade, I moved every thing to a new card.
My pi zero runs pihole + open vpn and is still on jessie. No reason to switch that one unless the foundation stops releasing updates for jessie.