Cassidy Dalva is Co-Founder and President at Miso Labs, a Y Combinator-backed AI company building the world's most emotive foundation models for voice.

Cassidy studied Economics and Digital Humanities at Stanford University, where she developed expertise in data science, multimedia journalism, and served as News Managing Editor at The Stanford Daily.

Cassidy is a multilingual communicator fluent in English, French, and Spanish, with journalism work featured in The Washington Post, the Los Angeles Times, and on Good Morning America.

At Miso Labs, Cassidy co-authored the MisoTTS research paper alongside co-founder Aoden Teo, introducing an 8-billion-parameter voice model achieving state-of-the-art emotive speech generation.

Miso Labs launched Miso One, the world's most emotive voice model with just 110ms latency, open-sourced its model weights on Hugging Face, and was selected for Y Combinator's Summer 2025 batch.