Generate Speech and Force New?

[joe_q_public]

Am I missing something? I am new to audtion so I fully accept that possibitity. I am having two issues with Audition's Generate Speech Function:

1. I am using largish text files, so each time I open the Generate window, it has to told the previous, completed text file into the editor, even though I am going to delete and add a new one. This is a laborious waste of time. Is there a way to a new Generator window with each open?

2. I frequently find that the AI does not translate a word or sound as expected, so I must edit it. If I open the Generate window it is going to reload the entire set text sequence, wasting valuable time, and dropping most of the effects I have applied to the file. Is there a way to edit the text without this happening?

I feel like I must be missing something obvious here. I am guessing the expectation is to not use large files but dozens of small files, but that just does not seem efficient. Though, granted, it anything would have to be more efficient to than the zero product after 3 hours of work I have at the moment.

Please advise... All comments welcome.

[SteveG_AudioMasters_ · Accepted solution]

The way text to speech works is completely out of Adobe's hands. It uses APIs built into the OS for the speech generation, and this offers very limited control indeed. It's slightly different on a Mac, I believe, but similar and still rather limited.

Everything we know about it (and personally I don't think it's very good at all) is here. Please note the comment about licensing for voices if you are using it for any commercial projects. I believe that for not much money, there are better systems around - a search reveals several.

[joe_q_public]

Thank you, sir  I had read the information you provided but didn't interpret it in proper context, i.e., that the function was solely reliant upon the functionality of the operating system. I appreciate the definitive statements.

joe

[c89]

Steve, your response makes no sense to me.  Adobe wrote the code to display the form that is displayed when Effect - Generate Speech is clicked.  Adobe could choose to clear the text box or provide an option to do so.  Current UI is very disappointing

[SteveG_AudioMasters_]

It's not as simple as that. If you look at what Narrator can and can't do, and how it handles things like text repeats, etc, you'll realise that it really isn't a suitable tool for the task in the first place - the idea of choosing to delete things doesn't feature in what is essentially a real-time app. Whilst it would be very easy for Adobe to make a much better working environment for it, I believe that it pretty much wouldn't work because of the command restrictions. If there was an easy way to fix this, it would have been done years ago.