Neural Codec Language Models Are Zero-Shot Text to Speech Synthesizers

Re:Neural Codec Language Models Are Zero-Shot Text to Speech... by kouteiheika on Friday January 06, 2023 @02:07AM (#34271830)

There is, just not for English.

Here, take a look at this snippet:

https://youtu.be/eEXvMOJ9ps0?t=66

They play 4 clips, 2 of them human, 2 of them AI generated. Can you tell which ones are which?

And the kicker is, this works in real-time (AFAIR), and it doesn't even use the GPU (it's CPU-only), and generates pitch-correct speech (for Japanese). It's not even funny how far ahead they are. And you can buy it right now.

AFAIK they use some sort of a hybrid method with a bunch of custom modeling DSP code around it (they've been doing speech synthesis for over a decade) plus a neural network. One mistake that essentially all of the western TTS models seem to make is that they use only a neural network, without augmenting it with non neural network code, which (from what I can see) is the secret sauce to make a fast and good sounding TTS work.

[ Parent | on HN ]

Re:Neural Codec Language Models Are Zero-Shot Text to Speech... by orbital-decay on Friday January 06, 2023 @02:36AM (#34271975)

Yes, but can it be used to express emotions? Can it derive emotions from the text alone, without the painstaking guidance? That seems to be the main culprit with existing TTS engines; neutral tone can be generated relatively well.

[ Parent | on HN ]

Re:Neural Codec Language Models Are Zero-Shot Text to Speech... by vanderZwan on Friday January 06, 2023 @07:44AM (#34274045)

Honestly I suspect some kind of "emotional markdown" would be more useful if it has a light and intuitive syntax

[ Parent | on HN ]

Re:Neural Codec Language Models Are Zero-Shot Text to Speech... by com2kid on Friday January 06, 2023 @10:26AM (#34275690)

Humans can't drive emotions from text alone without external contextual clues.

Such has been the source of much miscommunication online.

Amateur fiction writing, which tends to overemphasize how things are said ("I guess I can go rescue your cat", the exasperated detective said wearily) might be easier for AI!

[ Parent | on HN ]

Re:Neural Codec Language Models Are Zero-Shot Text to Speech... by Valgrim on Friday January 06, 2023 @04:02PM (#34281015)

Sure they can, that's the whole point of acting. Also anyone who ever read a story to a kid can infer emotions from the text itself.

[ Parent | on HN ]

Re:Neural Codec Language Models Are Zero-Shot Text to Speech... by com2kid on Saturday January 07, 2023 @08:25PM (#34295015)

To a limited extent, sure, but kids books are also written to be very emotive.

The linked page actually has examples of the same text being read with different emotions, demonstrating that for even a single sentence a lot of variance is possible.

[ Parent | on HN ]

Re:Neural Codec Language Models Are Zero-Shot Text to Speech... by burkaman on Friday January 06, 2023 @11:11AM (#34276426)

I can't, but I don't speak Japanese. Can a fluent Japanese speaker really not tell?

[ Parent | on HN ]