A simple wrapper around the SVox pico TTS pico2wave commandline tts tool for use in nodered on linux. This suite consists of one node that converts a string to spoken audio using svox pico tts.

To use this node you need to have the pico2wave libraries installed. On most systems with the apt package manager you can just use sudo apt-get install libttspico-utils to install it.

Unfortunately on Raspberrypi OS non-free Debian packages are not included by default. As an alternative you can download and install the packages manually as follows:

wget http://ftp.us.debian.org/debian/pool/non-free/s/svox/libttspico0_1.0+git20130326-9_armhf.deb
wget http://ftp.us.debian.org/debian/pool/non-free/s/svox/libttspico-utils_1.0+git20130326-9_armhf.deb
sudo apt-get install -f ./libttspico0_1.0+git20130326-9_armhf.deb ./libttspico-utils_1.0+git20130326-9_armhf.deb

or you can compile it from source.

To install the node execute npm install johanneskropf/node-red-contrib-pico2wave from your node-red folder (typically ~/.node-red). To install directly from a repository with npm you need to have git installed (sudo apt-get install git).

The node is very barebones right now. It accepts the text to be spoken as a string in the configured msg or context property or as a string directly entered in the nodes config screen. In the nodes config you can also set the language to be used (en-US, en-GB, de-DE, es-ES, fr-FR, it-IT) and if the node should output the tts audio as a buffer in the configured msg property or write it to a file on the file system. Either way the audio is output as mono 16000 kHz wav audio.