I am impressed with the Alexa / Sonos release yesterday and am pleased to see that it will develop. I know two other people with SONOS systems, and we all play music from a local NAS. Is this possible using Alexa voice control?
Page 2 / 4
I accept that this implementation is beta and has limitations.
I had already worked out that room-grouping isn't yet available by voice-control but that if I pre-group in the Sonos app I can use Alexa to play music on that group by sending it to any of the rooms in the group.
I had also discovered that, while I can't yet select music from Spotify using Alexa, if I add music from Spotify to a queue by using the Sonos app, I can control playback using Alexa
So far, so good BUT...
I am disappionted to learn that using Alexa to play from my local library is not considered a development priority; I have fairly eclectic musical tastes and while I find much to enjoy on Spotify, many of the albums stored on my NAS are not available there and although a fair number of them ARE available on Amazon, I don't feel inclined to repurchase as a digital download music that I already own.
I may be past MY prime but I am far from being a luddite and have embraced both digital downloads and streaming as enthusiastically as my grandaughters.
I am both a music fan and a musician who has been lucky enough to play music from most genres from the early rock era of the fifties right through to today's music and my listening enjoyment covers everything from classical, through early blues and jazz, 40's big bands (Ellington, Miller, etc), jazz combos such as Artie Shaw and Miles Davis, the 'crooners' - Crosby, Sinatra, Nat Cole and so on, plus all the big names that have come and some sadly gone (Elvis, Lennon, Bowie, Hendrix, Joplin) through the rock and pop era. There are not many notable names in any genre that are not represented somewhere in my collection.
While this Alexa integration does not prevent me playing anything in my collection via my beloved Sonos app, I'm sure that I'm far from being alone in hoping that Alexa voice-control will at some point be extended to include my hard-earned locally-stored collection.
I come from a generation that doesn't expect instant gratification, so I'm prepared to wait but PLEASE Sonos/Amazon don't relegate my personal collection to the back-burner. I'm sure I'm not alone....
You have described my situation to a "T". I have yet to order a Dot, although I have one teed up on amazon. Can you not add music to the Sonos Queue from your music collection, and play that queue with control by Alexa? I tend to play long playlists, often shuffled. Having the ability to control volume and turn off and on gets me pretty far along, although being able to select an obscure Sea Level or Johnny Winter track would be where I would like to wind up. I hope it is or becomes a development priority for all involved.
Yes, I can indeed both add to a queue from my NAS collection or select a playlist by using the desktop or phone/tablet app and then control playback using Alexa. However, listening to such a list often sparks a reminder of a song not currently in the list which I wish to add. At the moment I can't add a song 'on-the-fly' from either my collection or Spotify by using Alexa.
This is not a deal breaker, I can still use the app, but I'd been looking forward to not having to clean the bike grease from my hands (or dry my hands if I happen to be washing up) in order to do so....
(Sigh) I shouldn't complain of course, this is all a far cry from the fragile 78rpm records which formed part of my early listening experience or the ever-pulsing and fading reception of Radio Luxembourg....
This is not a deal breaker, I can still use the app, but I'd been looking forward to not having to clean the bike grease from my hands (or dry my hands if I happen to be washing up) in order to do so....
(Sigh) I shouldn't complain of course, this is all a far cry from the fragile 78rpm records which formed part of my early listening experience or the ever-pulsing and fading reception of Radio Luxembourg....
Any number of points to make in this conversation.
1. Alexa skills are not installed on any Alexa device (Echo, Amazon phone app, Amazon Music phone app, Eufy, Sonos 1), they live in the Alexa (Amazon) cloud and are attached to Alexa accounts, so updates will happen without any user intervention or opinion.
1a. This means the skills can get better as fast as the dev/qa/release process for the skill writer works.
1b. Stodgy old users like me need to get over the idea of staying on a specific version with specific functionality.
2. How to enable local library playback is not a hard problem, but there are interesting error paths that are more complicated. If the Sonos privacy policy allows replication of one's music library index to their cloud systems, then local library playback is easier to implement -- the alexa cloud just turns names into text and throws it at the sonos cloud which either plays the track or throws back an error. A little easier said than done, but @controlav's skill does that if one has a Win10 box up all the time on one's local network to interact with the local library via the Sonos SOAP API. Sonos's policy of not talking about roadmap and work-in-progress with any detail means we really don't know how close they are.
3. I suspect the Sonos point of view is that with modern streaming services have so much content that many *not all* local library users don't realize how much they have access to via those services. I've surprised myself this evening about how much of my local library is available, even if it doesn't play from my local source. There have been enough misses that I still really need/want local library playback, but I can see enough of a different point of view.
4. As i've said before in another thread, don't underestimate what's required in standing up the Sonos cloud resources involved in all of this. lechmere-v1.ws.sonos.com, the Sonos CA for Oath secure logins, the Akamai configs, the auto-scaling to provide a "presence" of every HHID that has v8 installed in the Sonos AWS accounts, and making that reliable at the uptime requirements for such a large user population. There's a huge amount of work involved. (And from what I can see, it's been done using the right pieces.)
1. Alexa skills are not installed on any Alexa device (Echo, Amazon phone app, Amazon Music phone app, Eufy, Sonos 1), they live in the Alexa (Amazon) cloud and are attached to Alexa accounts, so updates will happen without any user intervention or opinion.
1a. This means the skills can get better as fast as the dev/qa/release process for the skill writer works.
1b. Stodgy old users like me need to get over the idea of staying on a specific version with specific functionality.
2. How to enable local library playback is not a hard problem, but there are interesting error paths that are more complicated. If the Sonos privacy policy allows replication of one's music library index to their cloud systems, then local library playback is easier to implement -- the alexa cloud just turns names into text and throws it at the sonos cloud which either plays the track or throws back an error. A little easier said than done, but @controlav's skill does that if one has a Win10 box up all the time on one's local network to interact with the local library via the Sonos SOAP API. Sonos's policy of not talking about roadmap and work-in-progress with any detail means we really don't know how close they are.
3. I suspect the Sonos point of view is that with modern streaming services have so much content that many *not all* local library users don't realize how much they have access to via those services. I've surprised myself this evening about how much of my local library is available, even if it doesn't play from my local source. There have been enough misses that I still really need/want local library playback, but I can see enough of a different point of view.
4. As i've said before in another thread, don't underestimate what's required in standing up the Sonos cloud resources involved in all of this. lechmere-v1.ws.sonos.com, the Sonos CA for Oath secure logins, the Akamai configs, the auto-scaling to provide a "presence" of every HHID that has v8 installed in the Sonos AWS accounts, and making that reliable at the uptime requirements for such a large user population. There's a huge amount of work involved. (And from what I can see, it's been done using the right pieces.)
Before you spend any money, you can install either the Amazon app or the Amazon Music app on your phone and play around with it. You won't have the always listening aspect like an Echo or Sonos 1 -- you'll need to press something on the phone screen instead of saying "Alexa..." but you can speak every command and have it be processed just like an Echo would process it.
Likewise, I'm a little disappointed with the capabilities of the current Alexa integration. I have a comprehensive local music library, and don't have subscriptions to other music services. Its nice to see Sonos have released an Alexa skill , but for me its of limited use.
Before you spend any money, you can install either the Amazon app or the Amazon Music app on your phone and play around with it. You won't have the always listening aspect like an Echo or Sonos 1 -- you'll need to press something on the phone screen instead of saying "Alexa..." but you can speak every command and have it be processed just like an Echo would process it.
Thanks for the tip. The decision to purchase involves the confluence of a variety of tasks; sort of a "smart house" retrofit.
I get all the frustration, and it certainly is a downer if you don't want to use the services available to be controlled through Alexa. It's not bothering me really as I happen to primarily use Amazon, Pandora, and XM as my sources (other then tv line in), so I'm good. However, I get the frustration.
The way I see it, it's pretty clear that Sonos probably could have opted to design the interface the same was as the 'homebrew' skill. They probably didn't need to work with Amazon's developers hardly at all. If they did that then they probably would have had a lot fewer restrictions.
So what are some possible reasons why they didn't do that?
- Sonos wanted to have a device with Alexa built in. They obviously needed to work with Amazon on that. For design of the mics as well as using the service. I have to believe that that deal was tied to doing the speaker integration more along the way Amazon wanted it done. I doubt scraping their own Alexa enabled speaker was much of an option for Sonos, since their competitors were certainly going to go that route. Sonos can't afford to be behind on this.
- As has been pointed out in other threads, the vast majority of users use streaming services. I would also bet that the majority of people who have bought echo's primarily use the services Amazon provides. It does make sense that those services would receive the first treatment.
- There are some advantages to this design as opposed to Sonos 'going it alone'. Those being that you don't have to say 'tell sonos to' and eventually...hopefully...you won't have to specify the zone every time.
- Another advantage is the error handling. For example, If I were to say playing 'Play Tom Petty on Pandora in the Living Room', but I don't have a Tom Petty Station, Alexa will ask me if I want to create it. Other types of things. You can't do that sort of interaction with a homebrew skill.
- Then there is the visual and audio feedback. I the homebrew skill can reply (or could be programmed to) reply back that it's playing such and such in the living room, but it could give you the same visual feedback in your Alexa app, echo show and spots, or the fire tvs that are Alexa enabled.
To me, I can see how the homebrew design would have been easier to implement, and provide more for the customers initially. However, I can definitely see where doing the design this way provides for greater features in the future. Since the market is competitive and about to get even more competitive, I can see why they went this route.
Again, probably not much consolidation for those who can't use Alexa the way they want to now.
The way I see it, it's pretty clear that Sonos probably could have opted to design the interface the same was as the 'homebrew' skill. They probably didn't need to work with Amazon's developers hardly at all. If they did that then they probably would have had a lot fewer restrictions.
So what are some possible reasons why they didn't do that?
- Sonos wanted to have a device with Alexa built in. They obviously needed to work with Amazon on that. For design of the mics as well as using the service. I have to believe that that deal was tied to doing the speaker integration more along the way Amazon wanted it done. I doubt scraping their own Alexa enabled speaker was much of an option for Sonos, since their competitors were certainly going to go that route. Sonos can't afford to be behind on this.
- As has been pointed out in other threads, the vast majority of users use streaming services. I would also bet that the majority of people who have bought echo's primarily use the services Amazon provides. It does make sense that those services would receive the first treatment.
- There are some advantages to this design as opposed to Sonos 'going it alone'. Those being that you don't have to say 'tell sonos to' and eventually...hopefully...you won't have to specify the zone every time.
- Another advantage is the error handling. For example, If I were to say playing 'Play Tom Petty on Pandora in the Living Room', but I don't have a Tom Petty Station, Alexa will ask me if I want to create it. Other types of things. You can't do that sort of interaction with a homebrew skill.
- Then there is the visual and audio feedback. I the homebrew skill can reply (or could be programmed to) reply back that it's playing such and such in the living room, but it could give you the same visual feedback in your Alexa app, echo show and spots, or the fire tvs that are Alexa enabled.
To me, I can see how the homebrew design would have been easier to implement, and provide more for the customers initially. However, I can definitely see where doing the design this way provides for greater features in the future. Since the market is competitive and about to get even more competitive, I can see why they went this route.
Again, probably not much consolidation for those who can't use Alexa the way they want to now.
I can't see any motivation for Amazon to implement NAS, Spotify or Deezer "skills" with their Alexa integration, seems a conflict of interest. Since Sonos have teamed up and joint developed the One with Amazon I think we're out of luck with any none Amazon source.
Not a problem for me personally (I was weighing up buying my daughter a Sonos One for Christmas but the Alexa integration would be next to useless for us).
Not a problem for me personally (I was weighing up buying my daughter a Sonos One for Christmas but the Alexa integration would be next to useless for us).
Actually, the One is going to be platform agnostic. You will be able to select the voice assistant you wish to use, with Google being next on the slate.
It strikes me that some of the furor around all of the "new" stuff in the privacy update last month will be raised again, if indeed they need to upload in a cloud location all of the names of the music that people have locally. For me, it's not an issue, but then again, I've got a fairly high threshold for that kind of privacy issues. And frankly, Amazon already knows what CDs I've purchased from them. I don't consider my music to be a sensitive issue, like I do my bank account, social security, and other more pertinent data points. What I listen to isn't an issue, other than folks making fun of me. And I'm fairly inured to that, as well.
Plex stores the library in the cloud via an external connection to my nas. I can use my library on any machine and browser. I have in fact been using it this evening remotely in the U.K. While my nas is running in LA. Plex has an app for sonos which requires the Plex library to be accessible externally and not locally. This works flawlessly. So there should have been no reason not to implement this unless amazon wants the user to be forced to rent music through an amazon subscription. As is, Alexa on sonos only works fully if you are subscribed to an amazon rental prOgram. You cannot use full functionality otherwise
hmm are you referring to Plex via remote (as in make it viewable from outside your intranet)? which of course i have set up and use it with the Plex app and the website from work or wherever.
however i only have sonos at my apartment not any at my office. i should try to bring one in for a day and test that - no way am i leaving one here (no lock on my office door).
so if that is true than you are correct... there is no reason not to be able to access our local libraries on plex via sonos via alexa.
As far as I know, Sonos only supports Plex on a local server, not the cloud, and quite frankly, the Plex implementation is terrible and Plex basically abandoned it soon after it launched.
It is an interesting topic to ponder. Revealing the songs you collect in your library is probably a low risk privacy issue. Revealing how often you listen to certain tracks vs. other tracks could arguably be more of a concern. It is less about people making fun of you than it is a matter of people applying some statistical analysis like "people who listen primarily to this music genre are bad credit risks" or " they would not make good candidates to work here", etc.
Knowing in real time when you are at home listening to music is absolutely a risky piece of information to share. It can be used in a court of law to invalidate an alibi or used by thieves to establish a pattern for when your house is likely to be occupied or not.
It may sound far fetched but all of these scenarios have already occurred to some extent and are only expected to expand in the future.
Yeah, because hanging out down the street to watch you get in the car and leave is so much harder than hacking Amazon to figure out when someone isn't playing music. Jebus Christmas, do you people think about what you write? :8
Why would a thief waste time staking out a house when they can use data mining to get the info they need? Thieves are very sensitive to time management and return on investment.
Why would a thief waste time staking out a house when they can use data mining to get the info they need? Thieves are very sensitive to time management and return on investment.
Dear Lord, please tell me you are not being serious. :8
Why would a thief waste time staking out a house when they can use data mining to get the info they need? Thieves are very sensitive to time management and return on investment.
Dear Lord, please tell me you are not being serious. :8
I could say that all of my posts are absolutely dead serious... but that would by lying. 😃
Plex implementation is fine. Don't know what rubbish you are spouting about saying implementation is terrible. Works same or better than the crappy local
Library definition in sonos app, and at least we aren't crippled by the 65k music limit with the sonos default local library
Plex implementation is fine. Don't know what rubbish you are spouting about saying implementation is terrible. Works same or better than the crappy local
Library definition in sonos app, and at least we aren't crippled by the 65k music limit with the sonos default local library
Your opinion is decidedly in the minority. But to each his own.
Am I interpreting correctly what Alexa/Sonos integration can and cannot achieve?
I have just bought an Echo Dot and gone through the set up...
1 It CAN play a Radio Station in any of my Sonos devices, start and stop and control volume.
2 It CANT play anything from my local NAS ( this makes the whole thing useless to me)
3 It CANT play anything from Napster which is the only streaming service I have.
4 It CANT play anything from my Sonos playlist ( the playlists are mostly from the NAS, and some mixed NAS/Napster)
Can an expert confirm on 2-4 please.
The problem with this type of modern development methodology is that there isn’t a definition of what it should do? What I would call a Functional Spec. Where would I find this definition, as I’m not sure whether it’s because I don’t know how to get it to do something, or it just doesn’t do it!
I have just bought an Echo Dot and gone through the set up...
1 It CAN play a Radio Station in any of my Sonos devices, start and stop and control volume.
2 It CANT play anything from my local NAS ( this makes the whole thing useless to me)
3 It CANT play anything from Napster which is the only streaming service I have.
4 It CANT play anything from my Sonos playlist ( the playlists are mostly from the NAS, and some mixed NAS/Napster)
Can an expert confirm on 2-4 please.
The problem with this type of modern development methodology is that there isn’t a definition of what it should do? What I would call a Functional Spec. Where would I find this definition, as I’m not sure whether it’s because I don’t know how to get it to do something, or it just doesn’t do it!
It cannot initiate playback from anything except Amazon's music services, Pandora, SiriusXM, iHeart, and TuneIn, with Spotify coming soon. You can initiate playback of all other services via the Sonos app and then control playback pause, play, skip, volume, ask what's playing, etc. via Alexa.
See here for more details:
https://sonos.custhelp.com/app/answers/detail/a_id/4343?_ga=2.109578245.1024760305.1507348659-107072400.1489101718
Remember this Alexa implementation is currently in beta, and thus subject to change (hopefully for the better) as it gets closer to general release.
See here for more details:
https://sonos.custhelp.com/app/answers/detail/a_id/4343?_ga=2.109578245.1024760305.1507348659-107072400.1489101718
Remember this Alexa implementation is currently in beta, and thus subject to change (hopefully for the better) as it gets closer to general release.
The problem with this type of modern development methodology is that there isn’t a definition of what it should do? What I would call a Functional Spec. Where would I find this definition, as I’m not sure whether it’s because I don’t know how to get it to do something, or it just doesn’t do it!
Amazon is the owner of the API, which Sonos is first to use, but Denon and several of HK's other brands will soon integrate, as well. Amazon isn't sharing their specs publicly. Maybe if you ask them real nice...
Or, you can sign up for access here. Just be prepared to prove you're representing a company that's building smart speakers.
https://alexa.au1.qualtrics.com/jfe/form/SV_eJdai0Nm0gwLsQ5
OK thanks for the clarification - I’m afraid it’s pretty useless IMO, oh well put the Echo Dot on eBay......
That is a pretty fatalistic approach when something is still in beta and Sonos has stated it will be improving with every beta release. But to each his own.
Where the NAS play is concerned you can do all the subsequent things via voice for managing what is playing - volume/pause/stop/restart. So that is a plus over what you had before, so that is some use that wasn't there before surely? All you have to still do is initiate music play and if you use long playlists that remain in use over multiple listening sessions, that is a small task.
From what I have read elsewhere, given the architecture of Alexa+Sonos, it will be a long time before voice controlled local NAS play beyond that will happen - if ever it does. Pity, but local NAS users are also getting to be a minority in the user base so giving just them more than what they get today - which is a lot when one thinks about it - seems to be a remote possibility.
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.