Why we need evals for audio generative AI models? (opens in new tab)

(black.dubverse.ai)

5 pointspursuitcurves2y ago1 comments

1 comments

What criteria should we use to determine the best model when we have text-to-speech models such as ElevenLabs, Bark, etc?

How do we scale this up when these audio models have their "stable diffusion moment" (thanks simonw for the phrase).

j / k navigate · click thread line to collapse

(black.dubverse.ai)

5 pointspursuitcurves2y ago1 comments

What criteria should we use to determine the best model when we have text-to-speech models such as ElevenLabs, Bark, etc?

How do we scale this up when these audio models have their "stable diffusion moment" (thanks simonw for the phrase).

j / k navigate · click thread line to collapse