diajax 0.0.5a0

Dia-JAX: A JAX port of Dia, the text-to-speech model for generating realistic dialogue from text with emotion and tone control - this one to see if scan broken on colab tpu too

Dia-JAX

An experimental JAX port of Dia, the 1.6B parameter text-to-speech model from Nari Labs

Quickstart

Note: Currently only recommended for experimental/development use due to memory issues

output.mp3

pip install -U diajax
dia --text "[S1] Dear Jacks, to generate audio from text from any machine. [S2] Any machine? (gasps) How? [S1] With flakes and an axe. (chuckle) " --max-tokens=100

import diajax
model, config = diajax.load()
output = diajax.generate(model, config, text)
diajax.save(output)

Acknowledgments

This project is a port of the original Dia model by Nari Labs. We thank them for releasing their model and code, which made this port possible.

License

This project is licensed under the same terms as the original Dia model. See LICENSE for details.