HackerTrans
TopNewTrendsCommentsPastAskShowJobs

toebee

no profile record

Submissions

Show HN: Dia2, open-weights TTS model for realtime speech to speech

github.com
3 points·by toebee·قبل 8 أشهر·2 comments

Show HN: Dia, an open-weights TTS model for generating realistic dialogue

github.com
652 points·by toebee·السنة الماضية·190 comments

comments

toebee
·السنة الماضية·discuss
You're absolutely right. We used Jordan's Whisper-D, and he was generous enough to offer some guidance along the way.

It's also a valid criticism that we haven’t yet audited the dataset for existing list of tags. That’s something we’ll be improving soon.

As for Dia’s architecture, we largely followed existing models to build the 1.6B version. Since we only started learning about speech AI three months ago, we chose not to innovate too aggressively early on. That said, we're planning to introduce MoE and Sliding Window Attention in our larger models, so we're excited to push the frontier in future iterations.
toebee
·السنة الماضية·discuss
not part of the license!
toebee
·السنة الماضية·discuss
Thank you for the kind words! We only support English at the moment.. Hope to add more languages in the future.
toebee
·السنة الماضية·discuss
Thanks you!! We personally used Quickpod and Runpod the most. But you can try it now on HF Spaces without spinning up GPUs yourself!

https://huggingface.co/spaces/nari-labs/Dia-1.6B
toebee
·السنة الماضية·discuss
Thanks for the interest! We also enjoyed using E5-F2 :) You can try it now on HF Spaces: https://huggingface.co/spaces/nari-labs/Dia-1.6B
toebee
·السنة الماضية·discuss
Thank you so much for the kind words :) We only support English at the moment, hopefully can do more languages in the future. We are planning to release a technical report on some of the details, so stay tuned for that!
toebee
·السنة الماضية·discuss
We will try to make it work, but not sure if will be an easy task. For now, you can try with https://huggingface.co/spaces/nari-labs/Dia-1.6B
toebee
·السنة الماضية·discuss
Thank you! You can add audio prompts of calm voices to make them a bit smoother. https://huggingface.co/spaces/nari-labs/Dia-1.6B you can try it here!
toebee
·السنة الماضية·discuss
Thank you!! Works for English only unfortunately :((
toebee
·السنة الماضية·discuss
Thanks for the kind words! We're just following our interests and staying upwind.
toebee
·السنة الماضية·discuss
Thanks for the kind words! You can try it now on https://huggingface.co/spaces/nari-labs/Dia-1.6B Also, we'll try to update the Demo Page to something lighter when we have time. Thanks for the feedback :))
toebee
·السنة الماضية·discuss
Yes! But you would need to put together a LLM system that created scripts from the book content. There is an open source project called OpenNotebookLM (https://github.com/gabrielchua/open-notebooklm) that does something similar. If you hook the Dia model to that kind of system, it will be very possible :) Thanks for the interest!
toebee
·السنة الماضية·discuss
We're adding guides for Zero-shot voice cloning. You can try it using the second example on Gradio: https://huggingface.co/spaces/nari-labs/Dia-1.6B
toebee
·السنة الماضية·discuss
We just clarified in the README, sorry for the confusion ;(

Note that the model was not fine-tuned on a specific voice. Hence, you will get different voices every time you run the model. You can keep speaker consistency by either adding an audio prompt (a guide coming VERY soon - try it with the second example on Gradio or HF Space for now), or fixing the seed.
toebee
·السنة الماضية·discuss
Only two voices at the moment... We will need to upgrade the dataset to make that happen, and are considering that as one of the next steps.
toebee
·السنة الماضية·discuss
https://huggingface.co/spaces/nari-labs/Dia-1.6B we fixed it!
toebee
·السنة الماضية·discuss
You can try it now on https://huggingface.co/spaces/nari-labs/Dia-1.6B !!!
toebee
·السنة الماضية·discuss
Thanks for the kind words :)))
toebee
·السنة الماضية·discuss
We'll try to give a high-level overview when we publish the technical report!
toebee
·السنة الماضية·discuss
Thank you for the kind words! We don't have plans for that yet, but you can always open an issue or RP on Github.