SOFTWARE NEWS

Introducing Sonos Voice Control

Introducing Sonos Voice Control
Userlevel 5
Badge +2

Hi Sonos Community members!

Today, we introduce the first voice experience created purely for listening on Sonos. Designed with privacy at its core, Sonos Voice Control is the simplest way to control your music, offering complete command of your Sonos system using only your voice. 

Sonos Voice Control works on every voice-capable Sonos speaker, processing requests entirely on the device

No audio or transcript is sent to the cloud, stored, listened to or read by anyone.

Available on new voice capable products and as a free update for existing customers, Sonos Voice Control is compatible with Sonos Radio, Apple Music, Amazon Music, Deezer, and Pandora at launch. More services and markets will follow. 

 

Fast, accurate control of your music and your Sonos System 

Sonos Voice Control finds the music you want to listen to, processing requests entirely on the device. Local processing delivers faster response times, and effortless follow-ups. 

All you need is one “Hey Sonos” and you can follow up without the need for additional wake-words.

Just like the Sonos app, you can control music and speakers in any room, easily move music around the home, save and like your favorite songs to your personal music library and more.

 

Natural conversation with Giancarlo Esposito

Sonos Voice Control understands the nuance of human communication and will respond to natural commands like “turn it up!”. After a careful search, Sonos chose award-winning actor Giancarlo Esposito - best known for his roles in Breaking Bad, Better Call Saul and The Mandalorian - to deliver a familiar voice for US customers. With careful recording, advanced processing and mastering, the voice is natural, unobtrusive, yet confident and engaging. 

Sonos’ first voice will be joined by others over time as Sonos continues to expand the experience to new people and places.

 Sonos Voice Control is available in the US starting June 1 and in France later this year, with additional markets to follow. 

Read the full press release on Sonos Voice Control here.


149 replies

Badge +17

 

Very nice! Is it because SVC knows “cruise control” is a Sonos Radio station, or because of the keyword radio?  Asked a different way, does “hey Sonos, play Cruise Control” work?  Either way...really like this feature.

I’m assuming I can’t have a Pandora radio station, say ‘fleetwood mac radio’ and do the same thing. “hey Sonos, play Genesis” plays on Amazon and “hey Sonos, play fleetwood mac radio” plays on Pandora (it would play something on Amazon).  Instead I would need to  say ‘hey Sonos, play fleetwood mac on Pandora”...which wouldn’t be a big deal for me.

Yep, “Hey Sonos, play Cruise Control” works without the addition of “radio”. 

I don’t have Pandora to fully test this (since I’m in the UK), but “Hey Sonos, play Fleetwood Mac radio” played their Fleetwood Mac Mix Radio from Deezer on my system, with Amazon Music set as the default service. I would assume that a similar result would happen if Pandora was added to a system. 

If there are competing results where SVC can’t decide which one was intended, the speaker will prompt the user to select which one they’d like. For example, “Hey Sonos, play Back to Black by Amy Winehouse” could mean the song or the album. In this case, SVC would ask which one was intended.

Userlevel 5
Badge +14

Sorry for my ignorance, what does “ducking” mean?

Badge +17

Sorry for my ignorance, what does “ducking” mean?

Ducking means the speakers will lower (or “duck”) their volume to better hear the person speaking 🙂

Userlevel 5
Badge +14

Sorry for my ignorance, what does “ducking” mean?

Ducking means the speakers will lower (or “duck”) their volume to better hear the person speaking 🙂

Thank you, I obviously knew they did that but did not know that’s what it was officially called.

Also, with all the great questions, and as we get more answers I am wondering if it makes sense to create a more simplified, organized FAQ for Sonos voice control as a sticky?

 

Very nice! Is it because SVC knows “cruise control” is a Sonos Radio station, or because of the keyword radio?  Asked a different way, does “hey Sonos, play Cruise Control” work?  Either way...really like this feature.

I’m assuming I can’t have a Pandora radio station, say ‘fleetwood mac radio’ and do the same thing. “hey Sonos, play Genesis” plays on Amazon and “hey Sonos, play fleetwood mac radio” plays on Pandora (it would play something on Amazon).  Instead I would need to  say ‘hey Sonos, play fleetwood mac on Pandora”...which wouldn’t be a big deal for me.

Yep, “Hey Sonos, play Cruise Control” works without the addition of “radio”. 

I don’t have Pandora to fully test this (since I’m in the UK), but “Hey Sonos, play Fleetwood Mac radio” played their Fleetwood Mac Mix Radio from Deezer on my system, with Amazon Music set as the default service. I would assume that a similar result would happen if Pandora was added to a system. 

If there are competing results where SVC can’t decide which one was intended, the speaker will prompt the user to select which one they’d like. For example, “Hey Sonos, play Back to Black by Amy Winehouse” could mean the song or the album. In this case, SVC would ask which one was intended.

 

Again, very nice.  Seems like SVC tries to figure it out, rather than just doing nothing or something dumb.  

Thought this was interesting, although I want to hear more examples.

 

Userlevel 7
Badge +17

Sounds great! Now I hope that there will also be a version in Dutch!

Userlevel 5
Badge +14

One snippet of the response. I want more. I think that Sonos should Tweet out a snippet phrase a day.

And… I thought this engadget article was interesting, particularly the last couple paragraphs regard performance and choice of Esposito as the voice.  While I like the choice, it’s bold in the sense that Apple/Google/Amazon all went with a female voice that isn’t well known.  And there, is always the outside chance that Esposito does something that people don’t like, and no longer want to hear his voice.  True that you won’t hear his voice as much as other voice assistants, and I hope there are other voices available in English at some point.

Userlevel 6
Badge +17

Sounds great! Now I hope that there will also be a version in Dutch!

Misschien met de stem van Maarten van Rossum… 😏

Is it really only going to be launching in the US on 1st June, and not any other English speaking countries?

Looking forward to trying it in the UK.

Is it really only going to be launching in the US on 1st June, and not any other English speaking countries?

Looking forward to trying it in the UK.

 

Unfortunately, it’s not just about getting it to technically work in a language, but dealing with all the legalities and regulations each country has.

Userlevel 4
Badge +7

Designed long before there was knowledge about how such a system would potentially work. Seems like blaming Sonos for not allowing the PLAYBAR to have AirPlay 2. 

That’s a pretty I’ll informed response. It’s not like Sonos products didn’t already support voice assistants when the Five [and even the Play:5 (Gen 2] was released.

 

Pretty sure @Airgetlam is stating that the basic design of the play:5 and Five didn’t change, they only upgraded the memory other parts to modernize it.  Sonos didn’t change the speaker layout or the exterior housing to make room for mics and all that that entails...as well as testing, etc.  Whether the cost of additional development and such would have been worth it throw in mics, or what Sonos may have planned to address that in the future...I have no idea.

Also pretty sure the Play:5 (gen 2 was released years before Sonos One.  It certainly wasn’t after.

 

 

Fair call on the Play:5 (Gen 2), which was released about two years earlier than the One. However, given the shape and size of the Play:1 and One are similar, and they managed to squeeze in voice assistant support as part of that transition, I maintain there was no reason they couldn’t have accommodated the hardware for voice assistant support in the Five, and I said so at the time:

 

 

The One needed to have voice control.  No way around it in that case, since they needed a base/flagship device with voice control.  And you can’t assume that there was physical space in the Play:5/Five because there was physical space in the One/Play:1.  For one thing, the top of the One/Play:1 is flat, which works find for mics.  The top of the Play:5/Five is not, which probably isn’t good for mics.  They would have had to do more of a redesign.  I’m not say that the extra dev costs and such would definitely not be worth it, but way too much stuff unknown to me definitely say it was a stupid move.

Not sure why you feel the need to justify their choice? It was a dumb decision to not include assistant capabilities in the Five at the time it was released or since, and it is even dumber now that they are developing their own assistant.

Userlevel 3
Badge +2

While I know Spotify isn’t supported at the moment, I’m curious how it will work in my situation when it does . . . 

Our house has a Spotify Family account where each person has their own Spotify login. When Spotify is supported, will Sonos support multiple user ‘voice prints’ and map them to the appropriate Spotify account instances on the Sonos?  Sonos does already support having multiple instances of a service today, so it has at least part of the solution. And technically it could support this notion on any music service, not just Spotify.

 

This is highly desired so my Spotify account isn’t ‘polluted’ by what my wife and kids listen to 🙂 and to support multiple streams playing at the same time by different people through-out the house.

Google Home can do this via voice prints and auto-choose the right Spotify account (at least on Nest/Home devices, never tried on the Sonos). Alexa requires you to ask it to ‘switch profiles’ (and that does work on Sonos).

Is it really only going to be launching in the US on 1st June, and not any other English speaking countries?

Looking forward to trying it in the UK.

Interesting; I thought this would be baked into the version update to S2; updates that have been made available globally at the same time till now, as far as I know. Now this practice will change?

Userlevel 7
Badge +17

@Talvish That would mean the voice assistant would also need voice match (of which I have read nothing) and would not work in the same way the system does now, i.e. controlling the last used user profile for Spotify on this Sonos speaker. Though this would be very nice, I detect som hurdles here….

Badge +17

Is it really only going to be launching in the US on 1st June, and not any other English speaking countries?

Looking forward to trying it in the UK.

Sonos Voice Control will launch on June 1st worldwide, with US English being the supported language. This means you can enable and use it in the UK, but there may be accuracy issues with stronger accents.

 

While I know Spotify isn’t supported at the moment, I’m curious how it will work in my situation when it does . . . 

Our house has a Spotify Family account where each person has their own Spotify login. When Spotify is supported, will Sonos support multiple user ‘voice prints’ and map them to the appropriate Spotify account instances on the Sonos?  Sonos does already support having multiple instances of a service today, so it has at least part of the solution. And technically it could support this notion on any music service, not just Spotify.

 

This is highly desired so my Spotify account isn’t ‘polluted’ by what my wife and kids listen to 🙂 and to support multiple streams playing at the same time by different people through-out the house.

Google Home can do this via voice prints and auto-choose the right Spotify account (at least on Nest/Home devices, never tried on the Sonos). Alexa requires you to ask it to ‘switch profiles’ (and that does work on Sonos).

More services will be added to SVC over time 🙂

As for the voice prints/voice match, I’ve passed this feedback on to our development teams as a feature request.

Userlevel 7
Badge +18

Is it really only going to be launching in the US on 1st June, and not any other English speaking countries?

Looking forward to trying it in the UK.

Sonos Voice Control will launch on June 1st worldwide, with US English being the supported language. This means you can enable and use it in the UK, but there may be accuracy issues with stronger accents.


I can’t believe our British accents are worse than American ones 😉   Roll on, June 1st. 


I can’t believe our British accents are worse than American ones 😉   Roll on, June 1st. 

Surely it depends on what you mean here by British? 

Seriously though, I think that this feature as well as the fact that it will now roll out at the same time wherever a S2 Sonos product with a mic is working, is the most significant release under S2, never mind it took them 2 years after the S1/S2 event. Hi Res was a damp squib, as expected.

Is it really only going to be launching in the US on 1st June, and not any other English speaking countries?

Looking forward to trying it in the UK.

Sonos Voice Control will launch on June 1st worldwide, with US English being the supported language. This means you can enable and use it in the UK, but there may be accuracy issues with stronger accents.


I can’t believe our British accents are worse than American ones 😉   Roll on, June 1st. 

 

It’s science.  😎

Not sure why you feel the need to justify their choice? It was a dumb decision to not include assistant capabilities in the Five at the time it was released or since, and it is even dumber now that they are developing their own assistant.

 

I’m just clearing up some facts and pointing out there are a lot of unknowns and assumptions in your conclusion.  My conclusion was “I don’t know”.    

While I know Spotify isn’t supported at the moment, I’m curious how it will work in my situation when it does . . . 

Our house has a Spotify Family account where each person has their own Spotify login. When Spotify is supported, will Sonos support multiple user ‘voice prints’ and map them to the appropriate Spotify account instances on the Sonos?  Sonos does already support having multiple instances of a service today, so it has at least part of the solution. And technically it could support this notion on any music service, not just Spotify.

 

This is highly desired so my Spotify account isn’t ‘polluted’ by what my wife and kids listen to 🙂 and to support multiple streams playing at the same time by different people through-out the house.

Google Home can do this via voice prints and auto-choose the right Spotify account (at least on Nest/Home devices, never tried on the Sonos). Alexa requires you to ask it to ‘switch profiles’ (and that does work on Sonos).

 

Out of curiosity, is there alternative ways that this feature can be achieved without voice match?  Just wondering if adding the ability to set a default service by room/speaker would help to achieve the same thing.  You’ll almost always want the kids room to be  pointed to their account, right?  

Secondly, would it be useful to switch default account via voice command?  That would probably work ok for the wife, but not the kids if they were younger.

Userlevel 5
Badge +9

So I have Sonos Ones with Alexa.  Can someone explain why I would want to use this?  What are the advantages of this over Alexa?   I’m not sure what the benefits are over what we already have?

Also, will voice commands to set an alarm work?  Such as,  “Hey Sonos, set alarm for 7 AM”?   

So I have Sonos Ones with Alexa.  Can someone explain why I would want to use this?  What are the advantages of this over Alexa?   I’m not sure what the benefits are over what we already have?

 

 1  - It’s not an either or situation, you can have Alexa and Sonos voice on the same speaker

2 -  Privacy.  Alexa sends a recording of your voice to the cloud for processing.  Sonos does not.

3 - Speed.  Commands are prmised to process faster

4 -   Music service covered.   Right now, each have a subset of music services covered.  Sonos is promising to increase that.

5 -  Sonos voice allows you to group/ungroup speakers  at any time during audio playback.

6 - Based on discussion above, Sonos seems to be a little smarter in determining what music to  play than Alexa is.  Yet to be seen.

 

In my mind, it’s worth giving it a shot. If I don’t find it more useful than Alexa, I’ve lost nothing.

Userlevel 5
Badge +9

So I have Sonos Ones with Alexa.  Can someone explain why I would want to use this?  What are the advantages of this over Alexa?   I’m not sure what the benefits are over what we already have?

 

 1  - It’s not an either or situation, you can have Alexa and Sonos voice on the same speaker

2 -  Privacy.  Alexa sends a recording of your voice to the cloud for processing.  Sonos does not.

3 - Speed.  Commands are prmised to process faster

4 -   Music service covered.   Right now, each have a subset of music services covered.  Sonos is promising to increase that.

5 -  Sonos voice allows you to group/ungroup speakers  at any time during audio playback.

6 - Based on discussion above, Sonos seems to be a little smarter in determining what music to  play than Alexa is.  Yet to be seen.

 

In my mind, it’s worth giving it a shot. If I don’t find it more useful than Alexa, I’ve lost nothing.

Thanks for the reply.  Will Sonos Voice work with a music library on your PC?  Also, can you have it set an alarm?  

Reply