Synthesizing multiple lines often fails

[ronnystandtke]

I want to integrate your solution into an exam environment where students can input multiple lines of texts and get the text read aloud to them.

I noticed that synthesizing multiple lines of text often fails, e.g.:

I want to read
one line
and another.

The result is "stuttering" for many seconds on the second line. If the text is on a single line everything is fine.

The synthesized voice is absolutely brilliant in comparison to other solutions. If these little bugs could be ironed out, this TTS system would be almost perfect!

Thank you very much for your support.
Ronny

[synesthesiam]

I've pushed an update to master that adds an option to synthesize lines separately or as one. I haven't had time to re-build the Docker images yet, though.

You may also want to check out Larynx