3 ms·

I legitimately think this could be huge for self-published authors. It takes a skilled professional about forty hours of work to produce an audiobook from a nov

I legitimately think this could be huge for self-published authors. It takes a skilled professional about forty hours of work to produce an audiobook from a novel-length manuscript. Tacotron could do it in minutes.

jfengel 7y ago
I don't see that coming soon. The voice is one thing, but the performance goes far, far beyond that. Without understanding the text, you can't get good prosody out of a single sentence, much less developing a character for a whole performance. You'd have to "direct" this on a word-by-word basis: "Put the emphasis here. Speed up 10% here. Decrease vocal intensity 25%". You'd end up producing a whole "score", and it would take at least as long as the human actor puts into it. Having done that, it would be amusing to switch it from voice to voice, as a party trick. But the result would still be much poorer than you'd get out of an actor. Really solving the work of an actor is strong-AI-complete.
- jimmydorry 7y ago
  Feed in enough "good" audio books, and you would probably get something passable for smaller titles.
- sitkack 7y ago
  AI is separating the talent from the looks across the board. As it is now, one has to both be able to act and look good, but now the AI will enable those who can act, to be re-skinned, literally, into whatever the client needs.
- thomascgalvin 7y ago
  Audio books with genuinely good performances seem rare to me. There are a handful of voice actors that stand out, but many of the titles I've listened to have very flat delivery; the first sample in the original article has as much inflection as they do.
- animal531 7y ago
  What about using a tablet to direct the piece by drawing? You can get values for the intensity, speed and volume (up/down) pretty easily and intuitively. Even better if its linked to the voice generation system in real time, then you can save/redo sentences etc. as you go along.

jfengel 7y ago

I don't see that coming soon. The voice is one thing, but the performance goes far, far beyond that. Without understanding the text, you can't get good prosody out of a single sentence, much less developing a character for a whole performance. You'd have to "direct" this on a word-by-word basis: "Put the emphasis here. Speed up 10% here. Decrease vocal intensity 25%". You'd end up producing a whole "score", and it would take at least as long as the human actor puts into it. Having done that, it would be amusing to switch it from voice to voice, as a party trick. But the result would still be much poorer than you'd get out of an actor. Really solving the work of an actor is strong-AI-complete.

jimmydorry 7y ago
Feed in enough "good" audio books, and you would probably get something passable for smaller titles.
sitkack 7y ago
AI is separating the talent from the looks across the board. As it is now, one has to both be able to act and look good, but now the AI will enable those who can act, to be re-skinned, literally, into whatever the client needs.
thomascgalvin 7y ago
Audio books with genuinely good performances seem rare to me. There are a handful of voice actors that stand out, but many of the titles I've listened to have very flat delivery; the first sample in the original article has as much inflection as they do.
animal531 7y ago
What about using a tablet to direct the piece by drawing? You can get values for the intensity, speed and volume (up/down) pretty easily and intuitively. Even better if its linked to the voice generation system in real time, then you can save/redo sentences etc. as you go along.