search

readbeyond / aeneas

aeneas is a Python/C library and a set of tools to automagically synchronize audio and text (aka forced alignment)
http://www.readbeyond.it/aeneas/
GNU Affero General Public License v3.0
2.45k stars 218 forks source link

Technical description of method? #236

Closed ftyers closed 4 years ago

ftyers commented 4 years ago

First of all, thanks for what looks like excellent software, I found out about it after reading this overview.

I'm curious how it works? I see TTS is a requirement, does that mean, that you convert the transcript to speech and then try and segment by aligning the lengths of the passages ? Or is it more complex ? I'd love to see a paper if there is one, or just a more technical writeup!

chrisvaughn commented 4 years ago

Alberto has a good write up here https://github.com/readbeyond/aeneas/blob/master/wiki/HOWITWORKS.md

ftyers commented 4 years ago

That's perfect, thanks! Perhaps a link to that could go in the main README.md? Sorry for cluttering up your issue tracker!

chrisvaughn commented 4 years ago

Agreed. It's linked from the main project site https://www.readbeyond.it/aeneas/ under documentation but a lot of people start from the repo.

readbeyond commented 4 years ago

Thanks to both.

The "how it works" document is also linked in the main readme.md, in section "Documentation and Support":

https://github.com/readbeyond/aeneas/#documentation-and-support

AP

On 10/11/19 6:04 PM, Chris Vaughn wrote:

Agreed. It's linked from the main project site https://www.readbeyond.it/aeneas/ under documentation but a lot of people start from the repo.