SSML audio

[mohesles]

Ive been able to use SSML correctly except for audio tags.

I have tried hosted <speak>HI<audio src="https://s3-us-west-1.amazonaws.com/sayspring-prod/media/celtic-open-chime.mp3" />Welcome</speak>

and amazon native sounds <speak>HI<audio src="soundbank://soundlibrary/transportation/amzn_sfx_car_accelerate_01" />Welcome</speak>

In both scenarios the audio tags are just ignored by alexa and the rest spoken normally.

[thomas314159]

Thanks for the great node!

I also have no success with

[eRudy]

Hello,

same problem here: the audio tag produces no sound, even with the Amazon sound bank. Is this a limitation from Amazon ?

[586837r]

Don't know the reason but unfortunately audio tags don't seem to work at the moment.

[mohesles]

Don't know the reason but unfortunately audio tags don't seem to work at the moment.

Could it be that the code is escaping <><> tags?

[586837r]

No, any other tags work it is some limitation from amazon.

[martinschki]

second this...this would be really cool to use it as a door bell for example...

[zyrorl]

unfortunately it seems this is a limitation by amazon.. frustrating

[zimmer62]

Wondering if anything has changed since August... just wasted an hour trying to do this.

[tdejneka]

Just tried it today, sending to a 3rd generation Echo Dot using an Amazon account without Prime, and it was only able to produce the text portions of the SSML phrase, not the audio tags.

[877dev]

Same here, I realised SSML is the way to go but nothing works so far..

[877dev]

@martinschki recent update supports doorbell :)

[mfl76]

@877dev: Did you use the "Sound"-command or the "Speak" - SSML with audio tag? I did not get to work the <audio src="soundbank://soundlibrary/transportation/amzn_sfx_car_accelerate_01" /> in speak - ssml. The sound command works for me too.

[zimmer62]

soundbank didn't work for me also, but I do see some sounds now like doorbell.

[SteveRMann]

Amazon uses a subset of SSML. I use this as a guide:

SSML Tags Supported by Amazon Polly

[kfishermail]

I found sort of a work around to play a beep whenever something happens in my system. Use the "expletive" tag. Try this...

.

[mholttech]

So I was working on this just last night and if you look at https://developer.amazon.com/en-US/docs/alexa/custom-skills/speech-synthesis-markup-language-ssml-reference.html#audio it provides some information about how the MP3 must be encoded in a specific way. In addition to that I used this to play an MP3 file as a "Speak at Volume - Regular" command:

<speak> <audio src="https://jhsfkl78345hkjd.s3-us-west-2.amazonaws.com/alarm1-alexa.mp3" /> </speak>

[bbostock]

On a similar topic, I also want to make the 'Red alert' sound which works if I explicitly define it, but I can't get it to work from a message. Any ideas of the correct message format? This is what I use, and it returns success, but no sound. {"topic":"Alexa Message","type": "sound","payload":{"sound": "Red alert","devices": ["Kitchen Echo Show"]}}

[MTrinkiesJr]

Hey there, the solution from @mholttech works! The important thing is that the mp3 file is accessible from the internet via not self-signed https. Also you need to have a limit of 240 seconds, the bit rate must be 48kbps and the sample rate must be 22050Hz, 24000Hz, or 16000Hz. For more information to convert (even with audacity) see: 22050Hz, 24000Hz, or 16000Hz.

To trigger the mp3 sound from nodered use the "Alexa Routine" node. Use "Speak at Volume" with "Regular" option. For the text you can insert all mentioned here: eg:

`

Play sound

`

or only

`

[877dev]

I'm getting Alexa to say this at least, tried a correct format mp3 via github, dropbox, jovo tech and other websites in case it was the htpps issue...

"Sorry, I'm having trouble accessing your Simon Says EU skill right now."

[fpovoski]

Same problem as above "Sorry, I'm having trouble accessing your Simon Says EU skill right now."

[minhdirox]

Confirm the comment from MTrinkiesJr work! The format of mp3 files must be correct and the public URL endpoint must provide an SSL certificate signed by an Amazon-approved certificate authority.

[ronsegal]

This is an example that works for me:

<speak>
<audio src="soundbank://soundlibrary/musical/amzn_sfx_bell_timer_01"/>
<prosody rate="slow" volume="x-loud" pitch="-10%" intensity="medium">You have mail.</prosody>
</speak>