Hi everyone!
By way of I produce a twice-weekly, TTS-voiced, Linux and open source news podcast that aims not to waste your time. You can check out the latest episode here:
tabsnotspaces.com/shows/2020-1

And if you work on a project with a major milestone coming up, do get in touch so I can perhaps give it some publicity.
~ Paddy

@TabsNotSpacesAM I've subscried via RSS. I'll be sure to listen in. Our of curiosity what do you use for TTS because it sounds very close to real voice/speech. Only a few things where like "oh that sounded odd" and caused me to come back here to read TTS-voiced again then i was like "oh that makes sense now".

Thank you for your content.

@mgrondin I'm using one of Amazon's neural voices, with a small amount of post processing through ffmpeg to clean the audio up a little. Before settling on that engine, I tried literally everything else that I could find -- both open and closed source -- to see what worked best, with listenability my main criteria. Amazon's audio isn't the cleanest out there, but to my ears is far more human-like in how it handles intonation than any competitor. [1/2]

Follow

@mgrondin I'm afraid that the FOSS alternatives just aren't in the ballpark at the moment. But I'm not tied to my current implementation, and as I say on the website About page, quite prepared to 'fire' my current host as and when the technology improves ;) [2/2]

Sign in to participate in the conversation
Fosstodon

Fosstodon is an English speaking Mastodon instance that is open to anyone who is interested in technology; particularly free & open source software.