I use #ZenBrowser, since it's a fork of #Firefox, it has a [reader view](http://mzl.la/1cHarCP), which has a "Read aloud" mode

This allows you to read aloud the content of articles

This mode uses the [Web Speech API](https://developer.mozilla.org/en-US/docs/Web/API/Web_Speech_API/Using_the_Web_Speech_API#speech_synthesis), which is a standard API for web browsers

Sometimes, i want to test the #accessibility of content using a screen reader

I use #Linux, i use [Orca](https://orca.gnome.org) as a screen reader

The Web Speech API and Orca both use [speech-dispatcher](https://github.com/brailcom/speechd) to generate synthetic voices (Text To Speech AKA #TTS)

Speech-dispatcher can use several voice generators, i used to use [espeak-ng](https://github.com/espeak-ng/espeak-ng), which was the less bad

I played with espeak about 10 years ago, the voice was robotic and saturating

It was fun to play but unpleasant to use

Since then, there has been the boom of #BigData, Machine Learning and #AI

Now, espeak-ng seems to give me the same feeling as before

There are newer tools that generate voices than feel good, like [piper](https://github.com/rhasspy/piper), some voices are really great

You can [try listening to it here](https://rhasspy.github.io/piper-samples/)

For several months, i have been looking for a solution to plug Piper into speech-dispatcher

Today, i have found it !

Thanks to @mike, all i had to do was simply use [Pied](https://pied.mikeasoft.com)