Marine Audio Research

Project DoLittle

Whale bioacoustics, long-context models, and public release engineering.

DoLittle brings together CAIRN's public work on marine mammal communication: humpback whale datasets, long-context neural audio modeling, and the technical release materials needed to make the project legible to researchers outside the lab.

The project is intentionally split into two public surfaces: dataset work and model work. That makes the site more useful to readers who care about one side of the release more than the other, while still showing the broader research direction.

10Btokens in the humpback whale dataset release
3public model and methods release surfaces
128Kcontext scale in the longest published model track
DoLittle summary graphic
Release Detail

The detailed DoLittle public surface.

The public detail removed from the homepage belongs here: the scale of the dataset release, the model tracks, and the separation between dataset and model work.

Dataset Scale

The current DoLittle dataset release covers roughly 10B tokens of humpback whale song derived from NOAA SanctSound recordings.

Model Release Tracks

DoLittle currently exposes 3 public model and methods release surfaces so readers can separate architecture, training behavior, and release notes.

Long-Context Focus

The strongest published model track pushes to 128K context, which is central to the project rather than an incidental implementation detail.

Two-Part Project Structure

Dataset and model work are intentionally split into separate public sections so the project stays legible to technical researchers following one side more closely.

Dataset Release

Humpback whale dataset work.

The dataset side of DoLittle covers the construction of a large humpback whale training corpus from NOAA SanctSound recordings, including filtering, segmentation, tokenization, and release-facing documentation for long-context audio research.

DoLittle dataset pipeline
Pipeline

A concise view of the seven-stage data-processing flow behind the public release.

Dataset Release Blog

Overview of the 10B-token humpback whale dataset, the processing pipeline, and why the tokenized release is useful for long-context marine-audio experiments.

Read the dataset release

Repository

The public DoLittle codebase and supporting technical materials tied to the dataset and model releases.

Open the DoLittle repository

Hugging Face Dataset

The released humpback DAC9 corpus is also available on Hugging Face for direct inspection and dataset access.

Open the dataset page on Hugging Face

Model Release

Long-context marine audio models.

The model side of DoLittle covers training, checkpoint release, architecture choices, and the implementation details that matter when the problem is long-context neural audio rather than conventional short-window acoustic classification.

DoLittle long-context architecture
Architecture

One of the strongest public visuals for explaining why the model release is technically interesting rather than just biologically interesting.

Training and Release Blog

Detailed writeup of training runs, hardware constraints, loss behavior, and example generations from the DoLittle model family.

Read the model release

Architecture Notes

Technical notes on SWA+MoE and NSA+MoE design choices used to push context length while preserving a causal neural-audio decoder.

Read the architecture notes

Technical Release Notes

Supporting notes on the released checkpoints, configuration, and implementation details needed for reproducibility.

Read the technical release notes

Hugging Face Models

The released DoLittle model checkpoints are also available on Hugging Face for direct inspection and download.

Open the model page on Hugging Face