Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[JimH]

DESCRIPTION
Alexa is now an integral part of MC.

It requires Amazon access from the Internet.  More below.

It requires Media Server to be running and accessible. 

The Net Access topic on the wiki tells you how to do the two items above.

In the new Alexa Skill called JTalk, you will do the setup below one time.  You'll then have voice access to Media Center from a phone or other mobile device.  The access is similar to what you might have from any JRiver Remote, but the control is by voice.

SETUP
A phone or other mobile device that can run Alexa is required.  The mobile device is only necessary for setup. You can use any Alexa device after that.

INSTRUCTIONS
1) Make sure your MC 30 server is running and accessible from the Internet. Note that MC's built-in SSL will NOT work so you will have to make the non-SSL port available.

2) Open the Alexa app on your mobile device, go to Skills & Games, and search for JTalk.

3) Tap on the ENABLE TO USE button for JTalk.

4) When it shows the account linking page, type in the access key and credentials for your server and then tap Connect.

After setup, you can then issue voice commands like the sample below.  Voice commands will work on any Alexa device using the same Amazon account used for setup.

"Alexa, open j talk"
"play playlist random songs"
"next"
"pause"
"cancel"

SESSIONS
You open a command session by saying "alexa, open j talk" or "alexa, ask j talk to (specify command)".  You can issue multiple commands while the session is open.  You can close the session yourself by saying "cancel".  Otherwise, if you don't say anything for a while the session will automatically close.  This time is controlled by the specific Alexa device and is not currently editable.  You will have to open the session again to issue more commands.

EXAMPLE COMMANDS
play
pause
previous
next
stop
set volume to [3]
what's playing
list zones
set zone to [player]
play playlist [random songs]
start play doctor with artist [ziggy marley]
play artist [kenny loggins]
play album [martina]

You can also append desired zone and playback options to most commands:

play playlist [random songs] on zone [player]
play playlist [random songs] with options [shuffle]
play playlist [random songs] on zone [player] with options [add]

Complete Command List

JTalk Command Reference for Alexa

[JimH]

This is working now.

[kensn]

I have a large number of devices I use with Alexa. Some are retail devices, and some are homebrewed with ESP32 chips and such. I would really like to help in the testing and development if needed.

Ken 

[bytestar]

Not available in Germany?

[JimH]

Not available in Germany?

It looks like the Amazon Alexa products are available in Germany so this should work.

[Hendrik]

Skills are region limited. It needs to be specifically designed (and translated) for every country you want to offer it in.

[cncb]

For anyone that wants to test the beta, please PM me the email address associated with your Amazon account.

[marko]

Is it available for users in the UK?

[cncb]

Is it available for users in the UK?

It should be.

[JimH]

We hope to have this generally available very soon.

Name change:  JTalk

[Guernsey]

Sounds brilliant

I do not see the Skill "Media Center" in my skill search on my Alexa app on my tablet yet

Cheers Steve

[JimH]

Send a private message to cncb to ask to be in the beta group.

It will probably be in the Alexa store soon.

[marko]

Send a private message to cncb to ask to be in the beta group.

It will probably be in the Alexa store soon.

I did this yesterday, and, after a bit of to-ing and fro-ing, I logged into amazon.com with my amazon.co.uk credentials, which allowed me to add the skill, which shows up in the Alexa app on my phone as enabled, however, it does not work.

When I get to this step:

5) Type in: open media center

Alexa does not know what to do and replies asking if I actually meant "media server".

I guess this means it's not available in the UK.

-marko

[JimH]

Did you enable it?

[cncb]

When I get to this step:Alexa does not know what to do and replies asking if I actually meant "media server".

Please try again with: open j talk

When it's published (in the store, not in beta) it is supposed to be available in other English language locales.

[marko]

Did you enable it?

Yes

Please try again with: open j talk

When it's published (in the store, not in beta) it is supposed to be available in other English language locales.

Still nothing. See second attachment. Shame. Will wait patiently till it's available.

EDIT:
Jeez those attachments were HUGE! Fixed it now, is why I hate using my phone for these things

[JimH]

Are you using the Dev version?

[marko]

Yes. From the link Craig gave me last night. Is there another?

[JimH]

Not yet, as far as I know.

Open J Talk works for me now.

[marko]

Does case matter?

[JimH]

In searching in the app?  Don't think so.

Talking to Alex, I just say "open jtalk".

[JimH]

I'm talking to Alexa on an Echo device, not in the phone app.  Just checked a Dot and it works there also.

[JimH]

I just tried the Alexa app on an Android and I don't see a way to get it to work there.  It may only work on a device other than the phone.

[cncb]

I just tried the Alexa app on an Android and I don't see a way to get it to work there.  It may only work on a device other than the phone.

If you are typing it in you need to have a space: "j talk".  Otherwise, it works with voice on my phone.

[JimH]

Mark,
You're on 30.0.1.  I don't think it matters, but try updating.  You have to download manually once, then auto update will work again.

Also probably not related, but did you test access from the outside world?

Login / password? 

[marko]

This is all getting too confusing.

I am running 30.0.4
I use JRemote daily whilst driving. External access is all good.
I use a different port than the default. Assumed there would be a place in setup to specify that.
There is a login/password, but I can't get the skill to open in order to enter any details.

The launch phrase has changed to 'open j talk'?
The first post states that as a first time only thing, everything has to be typed, but later reading suggests this is no longer required??
Could the first post be updated with current instructions as and when they change?

It's flat-out not working for me. I'm guessing because I have a USA skill enabled on a UK account.

-marko

[JimH]

This is all getting too confusing.

I am running 30.0.4
I use JRemote daily whilst driving. External access is all good.
I use a different port than the default. Assumed there would be a place in setup to specify that.
There is a login/password, but I can't get the skill to open in order to enter any details.

The launch phrase has changed to 'open j talk'?

Yes.

The first post states that as a first time only thing, everything has to be typed, but later reading suggests this is no longer required??
Could the first post be updated with current instructions as and when they change?

Yes.

It's flat-out not working for me. I'm guessing because I have a USA skill enabled on a UK account.

Yes.  Probably.  We believe that will change when it's available through the Alexa store (not beta).  We don't know how long it will take.

Thanks for testing.

[cncb]

This is all getting too confusing.

Sorry for the confusion.  The only change is the launch phrase: open j talk.  You still have to do the setup once, please point me to where it suggests otherwise and I will fix it.

As Jim said, it looks like the beta is not accessible outside the US, but once it is accepted by Amazon and published, it is supposed to be available in other English-speaking locales.  Thanks for your help.

[cgott42]

Does it have an option to FF/RW the video by a variable (i.e. non standard) amount.
e.g. "Alexa FF 50 seconds"

[cncb]

Does it have an option to FF/RW the video by a variable (i.e. non standard) amount.
e.g. "Alexa FF 50 seconds"

It does not support seeking yet.

[dtc]

I cannot connect to my 2 servers. I have j talk enabled on my Android phone.  On my phone I type "open j talk" and then type "help" and get the help message. So, I am connected to the app.  But, when I do "set up access key xxxxxx" I get an error message "Sorry I have trouble doing what you asked. Please try again".  I have tried accessing 2 different servers, both of which Gizmo and JRemote can access from on the LAN and remotely and which can be accessed from MC on other Windows systems using the access keys.   The two Windows 10 systems are running .5 and .6. I have double and triple checked the access codes. I have tried this with the phone on the LAN and with the WiFi on my phone  turned off. Same result. This is in the US.   Probably something obvious - I just don't see it. Any suggestions?

EDIT : Media [url=http://wiki.jriver.com/index.php/Network_Access]Network Access[/url] Key - Test Access to this Server does not give an error. Authentication is turned off.

Working now. I needed the whole funky low, cap syntax for the access key.

[dtc]

First, thanks for doing this. It is a significant effort, but voice command is really the way of the future.

Here are some initial impressions and some issues.

The setup, especially the low, cap, num syntax is  is cumbersome. Hopefully that process can be made easier.

J Talk times out much too quickly. I am not sure if this can be controlled, but hopefully so. Otherwise, commands like Play can go to the default Alexa source, which is a pain.  Hopefully, that can be set to some long time.  Otherwise, the Tell J Talk will have to become the main usage.

The command list is very comprehensive and well thought out. Users will ask for tweaks, as they do to all of MC, but this is a great start.

I have tested many of the commands and most work correctly when typed in, but few work when spoken.  I have tried speaking them on both my phone (Android) and a Dot and  few other than Play and Help work when spoken.  Are others able to use the voice commands?  That may still be in development.

The Tell J Talk voice command makes you "announce" the command which is an extra step and seems unnecessary. Is this a debugging tool? Hopefully it is not a general feature.  The command should just execute.

Some issues, again all while typing commands.  I fully understand this is under development, so I expect problems like these, although some of them may be just me.

Higher is not recognized.  Lower works most of the time, but not always.  Quieter and Louder seem to work consistently.

Set Volume To a higher value only works some of the time.

Commands for View, Track, Album, etc. require the correct capitulation, which is understandable, but making them case free would be nice. Hopefully the voice commands will be case insensitive.

Play Artist "Artist" always starts at the same song.  That should be randomized.

It might be nice to have Unmute rather than using Play to unmute.  You can get use to this, but my first inclination is to unmute. Play seems to apply to what is playing not to volume.

Will continue testing.

Thanks.

[cncb]

The setup, especially the low, cap, num syntax is  is cumbersome. Hopefully that process can be made easier.

J Talk times out much too quickly. I am not sure if this can be controlled, but hopefully so. Otherwise, commands like Play can go to the default Alexa source, which is a pain.  Hopefully, that can be set to some long time.  Otherwise, the Tell J Talk will have to become the main usage.

I have tested many of the commands and most work correctly when typed in, but few work when spoken.  I have tried speaking them on both my phone (Android) and a Dot and  few other than Play and Help work when spoken.  Are others able to use the voice commands?  That may still be in development.

The Tell J Talk voice command makes you "announce" the command which is an extra step and seems unnecessary. Is this a debugging tool? Hopefully it is not a general feature.  The command should just execute.

Commands for View, Track, Album, etc. require the correct capitulation, which is understandable, but making them case free would be nice. Hopefully the voice commands will be case insensitive.

Play Artist "Artist" always starts at the same song.  That should be randomized.

Setup is a little messy, but you only have to do it once.  We might improve this in the future.

Unfortunately, we seem to have no control over the session timing out.  It apparently varies per Alexa device type.

You are getting unexpected results, because all the commands were tested via voice and should work.  What happens when you say "play playlist random songs"?  This should play the default smartlist "Audio -- 100 Random Songs".

I'm not sure what you mean by "announce" the command?  "tell j talk to play playlist random songs" will open the session and execute the command.

Case does not matter even when typing commands in.  I'm not sure why you are seeing this.

You have control over how things are played by adding the optional "with options" at the end of the command.  For example, "play artist sting with options shuffle" will do what you want.

[dtc]

I'm not sure what you mean by "announce" the command?  "tell j talk to play playlist random songs" will open the session and execute the command.

It asks me "Do you want to announce this command" or something like that. If I say yes it repeats the command, complete with my voice. It is expecting a Yes or No answer. If I say no, nothing happens. It may be an Alexa setting, but I have not heard it before.

I will do more testing of the voice commands. But most of them do not seem to work.  But now that I know they are supposed to work I will look further.

Thanks.

[cncb]

It asks me "Do you want to announce this command" or something like that. If I say yes it repeats the command, complete with my voice. It is expecting a Yes or No answer. If I say no, nothing happens. It may be an Alexa setting, but I have not heard it before.

I see that now.  When you type it in the command is executed as expected.  If you say "ask j talk to..." it will execute the command.  I will update the first post.  Thanks.

[dtc]

I see that now.  When you type it in the command is executed as expected.  If you say "ask j talk to..." it will execute the command.  I will update the first post.  Thanks.

It makes the Tell or Ask command very annoying, especially since it announces it on all Amazon devices that are connected. Any way to turn that off?

[cncb]

It makes the Tell or Ask command very annoying, especially since it announces it on all Amazon devices that are connected. Any way to turn that off?

It will only ask you if you want to announce with "tell j talk to...".  Use "ask j talk to..." to avoid the announcement.

[dtc]

It will only ask you if you want to announce with "tell j talk to...".  Use "ask j talk to..." to avoid the announcement.

OK, ASK works as expected and the commands work with ASK J TALK.  I think the Announce function with Talk was the reason I could not get the commands to work with voice. Thanks.

When I OPEN J TALK my dot only has an orange ring for about 5 seconds after I end giving commands. I thought J Talk was still connected, but apparently as soon as the orange ring goes off, it stops J Talk.  That means I have to use Open J Talk for almost any command. That makes that whole interface less useful.  But, with ASK maybe I will not need it.  Using ASK for everything also means I can use other Alexa commands while the music is playing.

I think my capitilzation problems may have been related to time outs. The commands seem case insensitive how.

[cncb]

When I OPEN J TALK my dot only has an orange ring for about 5 seconds after I end giving commands.

Interesting.  The session on my phone also stays open for only about 5 seconds but on my Echo Show it stays open for about 30 seconds.  It's too bad this is not a device setting for the user.

[JimH]

Interesting.  The session on my phone also stays open for only about 5 seconds but on my Echo Show it stays open for about 30 seconds.  It's too bad this is not a device setting for the user.

I've only seen about 5 seconds on an Echo and a Dot.

[dtc]

Interesting.  The session on my phone also stays open for only about 5 seconds but on my Echo Show it stays open for about 30 seconds.  It's too bad this is not a device setting for the user.

Perhaps a discussion of this issue is in order. If people know that the "orange ring" going out or reverting to Home on a phone cancels the skill, they can adjust their thinking as appropriate.  My phone also does about 5 seconds.

The ASK command seems to Open J Talk and leaves it open until it times out. Which means if you do 2 ask commands in a row, the second one errors since J Talk does not know what to do with ASK J Talk. Perhaps ASK should be a one time command that does not leave J Talk open.  That's at least how I interrupted it.  That is, use Open J Talk to open the skill and keep it open and use ASK J Talk to do a one time command.

[dtc]

I am not sure what Set Volume is doing. It uses a range of 1 to 10, which I would translate to 10% to 100%. However, when increasing the volume, it does not seem to do that. With the volume at 20%, using Set Volume to 10 only changes it to 25%. I would think it would set it to 100%. I am using Internal Volume and Maximum Volume is set to 100%. I can manually set volume to 100%.

EDIT: It looks like Set Volume to 10 actually adds 5% to the current volume, just like Louder.

Higher is not a recognized command.

[dtc]

What's Playing tells you the song that is playing

What Is Playing tells you the current zone rather than the song, I guess because What Zone Is Playing gives you the zone. It seems like if the word Zone is not present, it should tell you the song, not the zone.

[kevin_marchant]

Yes.Yes.Yes.  Probably.  We believe that will change when it's available through the Alexa store (not beta).  We don't know how long it will take.

Thanks for testing.

Long time UK-based J River user here (7+ years).  Alexa integration might be the key feature to get me to upgrade this time round.  The Alexa skill is still only available in the US Alexa store (see screen shots from the Alexa skill list on Amazon.com and Amazon.co.uk) so it looks like you pressed the "Certification" process on the developer page and it is now live - just not available in all regions.  I think UK English is treated as a different language to US English.

This might help - it talks about .. "you must deploy your AWS Lambda function code to multiple AWS regions"  https://developer.amazon.com/en-US/docs/alexa/smarthome/develop-smart-home-skills-in-multiple-languages.html

Hope you can get a wider deployment going for testing.

[cncb]

I am not sure what Set Volume is doing. It uses a range of 1 to 10, which I would translate to 10% to 100%. However, when increasing the volume, it does not seem to do that. With the volume at 20%, using Set Volume to 10 only changes it to 25%. I would think it would set it to 100%. I am using Internal Volume and Maximum Volume is set to 100%. I can manually set volume to 100%.

EDIT: It looks like Set Volume to 10 actually adds 5% to the current volume, just like Louder.

I cannot reproduce this.  "set volume to..." sets the volume for me as expected using Internal Volume or Application volume. Are you using some kind of digital output that might be causing unexpected behavior?

[cncb]

What Is Playing tells you the current zone rather than the song, I guess because What Zone Is Playing gives you the zone. It seems like if the word Zone is not present, it should tell you the song, not the zone.

Alexa handles all the voice processing, so we will probably have to remove the "what zone is playing" to fix it.  Likewise, we might have to change "higher" since it is not being recognized.

[cncb]

The Alexa skill is still only available in the US Alexa store (see screen shots from the Alexa skill list on Amazon.com and Amazon.co.uk) so it looks like you pressed the "Certification" process on the developer page and it is now live - just not available in all regions.  I think UK English is treated as a different language to US English.

Working on it.

[dtc]

Alexa handles all the voice processing, so we will probably have to remove the "what zone is playing" to fix it.  Likewise, we might have to change "higher" since it is not being recognized.

How about Which Zone is Playing?

[cncb]

How about Which Zone is Playing?

Yes, we can try that.  Thanks.

[dtc]

I cannot reproduce this.  "set volume to..." sets the volume for me as expected using Internal Volume or Application volume. Are you using some kind of digital output that might be causing unexpected behavior?

I am using a Chord usb DAC with WASAPI. Nice DAC but don't think it should be affecting this command.  It is a Chord driver.  I will try some other DACs.