Microphone sensitivity for the “Alexa” wake word is low

+1 here also. Would be good to get some certainty around a fix date as others have mentioned here.

I just can't get my head round this. The Echos and Dots, I believe, detect the wake word using the microphones and processing on the device itself (we are assured nothing is sent to the cloud until the wake word has been recognised and the lights come on) so I assume it's the same on the Sonos One. Surely Sonos tested the Sonos One in a few people's homes before releasing it to make sure that the microphones and the software, etc on the device worked. So how is it that lots of people are experiencing wake word detection to be so much worse than the Echo devices? And how can we be sure this can be fixed with only device software (not hardware) updates?

If we bought our Sonos Ones from Sonos then we have the option to return them in 100 days if this doesn't get better, but if we bought them elsewhere we may have only 30 days or less for this.
The listening is done at the player level of things, the microphones don't send anything up until the wake word is detected. However, software updates with the data of what the wake word sounds like go from the cloud to the player. Basically, the cloud trains the player on what to be listening for. That learning happens over time, and therefore it's a gradual process, one that's been taking place for over a year on Echo devices. There probably isn't going to be a single update that says "microphones fixed". They will get better over time.

The microphones on the Sonos One can be quite sensitive. I had an early test version that picked up just about anything with an "s" sound as a wake word. It's gotten a lot better. Picking out what is and isn't a wake sound is all about pinning down excess noise and figuring out accents, tone, and pitch, all while sorting through background noise.

Hope this microphone problem fixes soon, otherwise I will send back my Sonos One. My echo dot is so much better....

I'll pile on. I love how the sonos one sounds, but the mic is the worst. If I knew for a fact this was going to be fixed in a week or two I'd hold on to it. As it is I'm returning it for an echo. Dumb me thought at least the voice recognition would be decent so I splurged for the better sound.

+1 to this frustration. After using an Echo for over a year, it's annoying to have to yell at the Sonos One. I've bought two for a small room, and at 2 meters away the One will still have a 50% wake rate.

Really looking forward to the software update, but also considering returning until that time.

Also experiencing the need to speak a little louder, but it's not a showstopper.

What's more annoying to me is the "I'm listening" beep. Way too loud for my taste, and it gets even louder when the music volume is turned up. A setting to adjust that beep volume would be appreciated.

I don't know that this is strictly a sensitivity issue -- after some more testing, I can stand 3 feet away from the ONE, without changing pronounciation, volume, or tone, and say Alexa multiple times. It will pick up, at best, every other "Alexa". I expected the speak to play features to be beta, but just hearing the wake word is the most core of features the ONE is purchased for!

So, are we talking days, months, years? Will the improvement be before the 100 days return window expires? I'm close to sending 3 Sonos Ones back now.
Though the microphones work great in the majority of homes and setups, we're at that point where we're collecting more and more samples and fine-tuning the microphones. Voice control improves in stages, so I'd expect to see improvements as releases come. There isn't just going to be one update, but several that continue to tine the microphones and the Sonos One in general. One top of the microphones, you'll see the Alexa skill get better too, improvements on that may be as quick as every couple weeks, you might not notice what's changed though, depending on what it's targeted at. Those updates would largely be handled on the cloud side of things, so it wouldn't be a software update on your system (those will come too though).

My suggestion would be to keep using the Sonos One. If you do have a lot of trouble with it hearing you, see if there's anything that might be causing a lot of background noise right next to it. That would help in the short term.

Just set up my Sonos 1 speakers, I was actually looking at how to change the Wake word since it is different on my other units but looking at the answers here I think I will leave the speaker's wake word set to the default Alexis and leave my other units set to a different wake word, then they won't get confused about what system I am speaking to

Will the improvement be before the 100 days return window expires?
Now that is what I would call an excellent question.

Thanks for sharing everyone. The team is constantly working with gathered data around the microphones as we tune the Sonos One and Alexa to improve the microphone’s ability to pick out just the right sounds. This’ll get better over time as Alexa and the Sonos One learns.
So, are we talking days, months, years? Will the improvement be before the 100 days return window expires? I'm close to sending 3 Sonos Ones back now.

Thanks for sharing everyone. The team is constantly working with gathered data around the microphones as we tune the Sonos One and Alexa to improve the microphone’s ability to pick out just the right sounds. This’ll get better over time as Alexa and the Sonos One learns.