It is very common to run into translation needs when text input is not possible. Few examples,
Phone/Internet bill in pdf
Bank documents in pdf
Something in a grocery store (real-life) when taking an image is the most convenient
German content as online images (which you don't want to type)
In the remainder of this issue, I will refer to the input for all of the above use cases as "document(s)".
The echoing superpowers of Tuti can be used to help here! The way I imagine this feature is the following:
Using telegram (current medium) as an interface to send documents to Tuti
Tuti translates them back in the same format or text and reverts
Step two can be separate commands for either annotated response or text response
Surely, text response is relatively easier to build but annotated response has its valid use cases. Specially, when you are about where in a document is something written.
It is very common to run into translation needs when text input is not possible. Few examples,
In the remainder of this issue, I will refer to the input for all of the above use cases as "document(s)".
The echoing superpowers of Tuti can be used to help here! The way I imagine this feature is the following:
telegram(current medium) as an interface to send documents to TutiSurely, text response is relatively easier to build but annotated response has its valid use cases. Specially, when you are about where in a document is something written.