Is there a way to pull a load of recordings?
I am working on a school project at the moment and I’ve chosen to test the accuracy of speech recognition systems.
How can I download 50 unique reference recordings? (For example)
We haven’t published the data yet, but will later this year.
In the meantime, here is a large collection you could use:
I’ve downloaded the corpus you link to.
It doesn’t look like the text which is being read is stored alongside the audio data.
This is quite important to me so I can check the accuracy.
I’ve haven’t looked into it, but I know the text is there somewhere, you just might need to write a script to format the data the way you want it.