OCR + quick voice reading [Feature Request]

Supaiku

举人
It would be amazing if you could have a button which would pause the OCR and read the text real quick, and then unpause and continue. So that you could effectively read a book aloud with the app with only a tap every time the characters are correctly recognized.

I hope your team will be able to add this feature (or some variation thereof) :)
Thank you
:)
 

mikelove

皇帝
Staff member
Supaiku said:
It would be amazing if you could have a button which would pause the OCR and read the text real quick, and then unpause and continue. So that you could effectively read a book aloud with the app with only a tap every time the characters are correctly recognized.
Do you mean that it would just read the recognized word? Our newly-licensed text-to-speech system should be able to facilitate this for full pages of text in the document reader, so it seems like that might be a better fit for your purposes - read a whole paragraph of captured text aloud instead of just one word.
 

Supaiku

举人
I mean, in the realtime OCR mode, read whatever characters are recognized (in paused state, when fixed, or on button press).
It could be a little unweildy... I dunno - it could work well though.
The trick is that it would be reading a physical, paper book, not a digital book:) (but it's great to know about the read aloud/reader compatibility! :)
 

mikelove

皇帝
Staff member
Supaiku said:
I mean, in the realtime OCR mode, read whatever characters are recognized (in paused state, when fixed, or on button press).
It could be a little unweildy... I dunno - it could work well though.
The trick is that it would be reading a physical, paper book, not a digital book:) (but it's great to know about the read aloud/reader compatibility! :)
We've actually had a few requests for something like this on iOS too... doesn't even have to be tied to a button press, we could theoretically trigger it if you just point at a word for long enough (we already have a flashcard "barcode reader" mode on iOS that works that way).
 

Supaiku

举人
mikelove said:
Supaiku said:
It would be amazing if you could have a button which would pause the OCR and read the text real quick, and then unpause and continue. So that you could effectively read a book aloud with the app with only a tap every time the characters are correctly recognized.
Do you mean that it would just read the recognized word? Our newly-licensed text-to-speech system should be able to facilitate this for full pages of text in the document reader, so it seems like that might be a better fit for your purposes - read a whole paragraph of captured text aloud instead of just one word.
I just tried it in reader and noticed that reading more than just the highlighted text is not yet an option. Is this something we can hope for sooner, rather than later?
 

mikelove

皇帝
Staff member
Supaiku said:
I just tried it in reader and noticed that reading more than just the highlighted text is not yet an option. Is this something we can hope for sooner, rather than later?
Relatively soon, we hope - the coding part of it is easy, but we still haven't come up with a UI that we're happy with for longer text reading.
 
Top