Moshi: a speech-text foundation model for real-time dialogue

Full-duplex spoken dialogue — both voices as parallel token streams, with a 200ms inner monologue.

Défossez et al. · arXiv 2024 · Foundations. Read the paper ↗

A free, interactive, animated visual explainer of Moshi: a speech-text foundation model for real-time dialogue — every exhibit computed from the real formulas, with verbatim quotes from the source.

Questions

What is Moshi: a speech-text foundation model for real-time dialogue?

Full-duplex spoken dialogue — both voices as parallel token streams, with a 200ms inner monologue.

Who published Moshi: a speech-text foundation model for real-time dialogue, and where?

Défossez et al. — arXiv 2024 (arXiv:2410.00037).

Where can I find a visual explainer of Moshi: a speech-text foundation model for real-time dialogue?

Right here — a free, interactive, animated walkthrough of the whole paper, with exhibits computed from the real formulas and verbatim quotes from the source.

Related explainers

• Attention Is All You Need
• GPT-3: Language Models are Few-Shot Learners
• Mixtral of Experts
• Training Compute-Optimal Large Language Models
• Mamba: Linear-Time Sequence Modeling with Selective State Spaces
• BERT: Pre-training of Deep Bidirectional Transformers
• Scaling Laws for Neural Language Models
• Adam: A Method for Stochastic Optimization