Microsoft VibeVoice: A Frontier Open-Source Text-to-Speech Model

Re:Microsoft VibeVoice: A Frontier Open-Source Text-to-Speec... by echelon on Wednesday September 03, 2025 @09:44AM (#45115714)

This is close to SOTA emotional performance, at least the female voices.

I trust the human scores in the paper. At least my ear aligns with that figure.

With stuff like this coming out in the open, I wonder if ElevenLabs will maintain its huge ARR lead in the field. I really don't see how they can continue to maintain a lead when their offering is getting trounced by open models.

[ Parent | on HN ]

Re:Microsoft VibeVoice: A Frontier Open-Source Text-to-Speec... by kamranjon on Wednesday September 03, 2025 @10:26AM (#45116221)

Hmmmm… what is your opinion on the examples showcased here vs the ones on the Dia demo page?

https://yummy-fir-7a4.notion.site/dia

I am not sure why but I find the pacing of the parakeet based models (like Dia) to be much more realistic.

[ Parent | on HN ]

Re:Microsoft VibeVoice: A Frontier Open-Source Text-to-Speec... by watsonmusic on Wednesday September 03, 2025 @10:58AM (#45116592)

11labs is facing a real competitor

[ Parent | on HN ]