Missing
Fundamental
LabsGet in touch
Independent lab / Open audio research001 — A different scale

Missing
Fundamental
Labs.

We study what it takes
to make small audio models
sound fully human.

The note you hear. The frequency we leave out. f₀ absent
0 ms200 ms
2f₀ + 3f₀ + 4f₀ + 5f₀ + 6f₀Synthetic signal.
More from less.

Small models.
Open questions.
Shared progress.

Meet the lab
Speech synthesis · Efficient models · Open scienceExplore our work
01 / Research

Work in the open.

All research
Text-to-speech / Project 001RivetA voice. A smaller footprint.
In development

A small model.
A voice of its own.

Rivet is our core text-to-speech project. We’re exploring expressive speech with a focus on model size, practical inference, and reproducible methods.

Explore Rivet
Lab note

A different scale

Why we’re starting with smaller models, and what we want to make open.

Lab note

The note that isn’t there

The auditory phenomenon behind our name: a pitch can survive the absence of its fundamental frequency.

02 / Our approach

Research is more useful
when you can build on it.

Our aim is to share the methods, evidence, and limitations alongside the models, so others can understand and reproduce the work.

What we believe
03 / Data

Know what
goes in.

Provenance, preparation, and evaluation.

04 / License

Know what
you can build.

Clear terms, attached to every release.