[PLUGIN] Text-to-speech

Hi there,

I’ve started creating a plugin for text-to-speech using Amazon Polly.
Current features include automatic track announcement, a socket.io listener for messages to be read out loud (‘tts-say’ with payload {“text”: “…”}) and one for announcing the current track (‘tts-announce’). Works good with node-red.

Install will take a while, some dependencies need to be build.
You’l need an AWS API key and secret, enter them in the plugin settings page.
Voice is configurable, see API docs for valid ids.

Make sure you’re able to play multiple simultaneous audio sources, for me this config works (Odroid C1, usb soundcard):
/etc/asound.conf

[code]pcm.custom
{
type plug
slave
{
pcm “dmix:2,0”
}
}

ctl.custom
{
type hw
card card2
}[/code]
/etc/mpd.conf

... audio_output { type "alsa" name "alsa" } ...

Download
github.com/ivesdebruycker/volum … _0.1.0.zip

Development
github.com/ivesdebruycker/volum … -to-speech

Well done!!!

Few suggestions:

  • When you need to build modules, is better to include them already built in the .zip file, or download their binaries. You can see an example on how we deal with them in the spop install script github.com/volumio/volumio-plug … install.sh
    this way we download the zipped binaries for the right platform.

  • Is there any way to make it more "Plug and play?)

Hello Ives,

can you tell me, how you connect volumio in node-red?
When i write the IP of volumio in the websocket-listener, I become a Error: ws:undefined : Error: not opened