Adobe recently announced Project VoCo — an experimental tool that will allow users to remove and replace certain words in a speech. It even includes adding words that did not initially appear in the audio file.
.. But I like the whole idea!!
According to TechCrunch, “Project VoCo needs about 20 minutes of voice samples from a given speaker. It then analyzes the speech, breaks it down into phonemes, transcribes it and creates the voice model.”
Unfortunately, this is a prototype and no information on if or when Project VoCo will become a commercially available product.