Does anyone here know of a good text-to-speech engine? I'm trying to render Swedish. I've played with espeech, Google's TTS and Amazon Polly. It turns out that GTTS with WaveNet voices is magically good, Polly is good enough, while espeech sounds like my Amiga did back in the late 80s/early 90s...
Is there any FOSS alternative out there that actually works?