Truncated text

TTS (Text-to-Speech)
#1

Hi
I’m trying out the TTS for rather longer texts, but it seems text gets truncated. Is there a specific configuration that I need to set for that?

For example, a news article with 17 lines of text is truncated so the wav speech file only reads the first 5 lines.

I’m using the Docker container packaged by Michael Hansen at https://github.com/synesthesiam/docker-mozillatts .

#2

In it’s current phase it is best to generate just one sentence at a time and connect them manually.

#3

Thanks for the tip. I’ll try implementing a queuing and aggregation mechanism for it.

#4

You may want to look into the server implementation (dev branch). There is a simple sentence chunking already implemented.

2 Likes