Sonos TTS Voicers API - delay in hearing voices

[AutoFrank 372]

Hi,

I'm using the 1.0.1b beta of the Sonos remote with the voicers API.

 

I have a basic test scene to develop voice usage in my system

 

From when I start the it takes between 8-9 seconds before I hear the sentence. The sonos mutes from when I start the scene

 

The code is a simple function

Please login or register to see this code.

Please login or register to see this code.

The actual code to "speak" is separate as this allows me to call it with the required sentence and not repeat all the code

 

The TTSEngineID code is from the forum here

Please login or register to see this code.

Is there anybody else experiencing this type of delay (or perhaps its normal ??)

Is I used MP3's on my local NAS would it be faster?

 

Thanks

-F

[Guest 22]

I'm using "fixed" MP3's on my NAS for this kind of things!

Way faster.

 

I'm using

Please login or register to see this link.

to create the MP3's (you will need to edit / cut-off the last part of the MP3)

[AutoFrank 372]

I'm using "fixed" MP3's on my NAS for this kind of things!

Way faster.

 

I'm using

Please login or register to see this link.

to create the MP3's (you will need to edit / cut-off the last part of the MP3)

 

Thanks Guest, I'll take that route as well.....

 

Could you share the code you use for the MP3 approach ?

What app do you use to trim the last part of the MP3 /

 

-F

[AutoFrank 372]

I'm using "fixed" MP3's on my NAS for this kind of things!

Way faster.

 

I'm using

Please login or register to see this link.

to create the MP3's (you will need to edit / cut-off the last part of the MP3)

 

Just found this one 

Please login or register to see this link.

 

There is a good choice of voice types and nothing to trim at the end.......

Please login or register to see this link.

[Guest 22]

Nice find, but I need Dutch, and fromtexttospeech does not do this

Please login or register to see this image.

/emoticons/default_sad.png" alt=":(" srcset="https://forum.fibaro.com/uploads/emoticons/[email protected] 2x" width="20" height="20" />

 

Will stick with text2speech for now.

For the records.. I'm using Audacity to edit the MP3's (see:

Please login or register to see this link.

 for more info)

[AutoFrank 372]

Nice find, but I need Dutch, and fromtexttospeech does not do this

Please login or register to see this image.

/emoticons/default_sad.png" alt=":(" srcset="https://forum.fibaro.com/uploads/emoticons/[email protected] 2x" width="20" height="20" />

 

Will stick with text2speech for now.

For the records.. I'm using Audacity to edit the MP3's (see:

Please login or register to see this link.

 for more info)

 

Ah... now I get your avatar tagline

Please login or register to see this image.

/emoticons/default_icon_smile.gif" alt=":-)" />

 

do you have a code snippet to share that I could use to get it to play MP3 instead of voicers..

[Guest 22]

Sure..

 

1st my doorbell scene.
 

Basically it waits for value 152 to change (the doorbell button). When so, it will populate the global variable MP3_Sonos to the needed MP3 to be played.

After this it will push the "play" button in a VD to "play" it.

Please login or register to see this code.

My VD to actually play the populated MP3_Sonos global variable looks like this. (code is within a button of the VD)

Please login or register to see this code.

[AutoFrank 372]

 

Sure..

 

1st my doorbell scene.

 

Basically it waits for value 152 to change (the doorbell button). When so, it will populate the global variable MP3_Sonos to the needed MP3 to be played.

After this it will push the "play" button in a VD to "play" it.

Please login or register to see this code.

My VD to actually play the populated MP3_Sonos global variable looks like this. (code is within a button of the VD)

Please login or register to see this code.

 

 

Thanks - That'll help a lot

 

much appreciated

-F

[AutoFrank 372]

Thanks - That'll help a lot

 

much appreciated

-F

 

 

 

Sure..

 

1st my doorbell scene.

 

Basically it waits for value 152 to change (the doorbell button). When so, it will populate the global variable MP3_Sonos to the needed MP3 to be played.

After this it will push the "play" button in a VD to "play" it.

Please login or register to see this code.

My VD to actually play the populated MP3_Sonos global variable looks like this. (code is within a button of the VD)

Please login or register to see this code.

 

Hi Guest,

I did a quick test last night and it works very well and the latency is down to a 1-2 seconds

 

BTW - I notice that the Sonos x-object variable seems to be getting corrupted a bit. I'm not sure if its related to the TTS work or not

Do you notice any issue with your implementation ?

 

-F