given the hype of so-called artificial intelligence, I wonder, having been a long time user of audacity to create teaching content that I use mainly in H5P presentations, if it might be worth considering the following.
An audio file containing many spoken words separated by two second pauses. These should be provided with a text track that gives them a name that says exactly what they contain.
For example: audio=tree; text track=tree etc. The aim is therefore to assign names to the individual parts of such relatively large audio files so that they can then be used for further processing in other content via mp3 export.
That sounds like voice-cloning, which is available now …
I may have expressed myself a little awkwardly and misleadingly, but I meant an existing file with many already generated spoken words, with the respective word name in the text track to then export them as an mp3 to be able to use them in my learning content.
Create a label track, then export-multiple based on those labels. That will create individual audio files of each labelled section with your label-text as the audio file-name.
[Converting speech to text via AI is now possible, (but not in Audacity) ].
This topic was automatically closed after 30 days. New replies are no longer allowed.