FOR AI TEAMS
Human conversation.
More context for AI.
Explore English recorded conversations with audio, video, and transcripts. Start with the source material, then define the collection your model needs.
CURRENT COLLECTIONS
Get close to the data.
Long-form English speech and conversation, with collection-specific specifications you can inspect.
English conversations
Browse the full collection of English podcast audio, video, and transcripts.
Two-speaker conversations
Back-and-forth dialogue, with individual speaker tracks where recorded.
Solo speech
Single-speaker recordings for long-form speech and transcription tasks.
THE CONNECTIONS MATTER
A recording.
And everything
around it.
Evaluate the camera views, audio tracks, transcript options, and source context together. We distinguish recorded assets from annotations that are available or would need to be commissioned.
Explore the recording structureA CONSIDERED PROCESS
Specific from the start.
A sample, a clear scope, and a documented delivery.
Start with the task.
The model, the recording conditions, and the decisions it needs to make. Define what a useful sample would show.
Evaluate the source.
Inspect representative media, speaker structure, transcripts, and technical specifications before defining a delivery.
Agree the handoff.
Confirm the permitted use, selected assets, annotations, and documentation for the exact collection.
CONTENT TYPE GUIDES
Understand the possibilities.
Explore media categories, their applications, and what makes them useful for AI. Visit the catalog to discover our collections.
Conversational speech
Natural, unscripted, multi-speaker dialogue — interviews, calls, and conversations recorded in real conditions.
Read speech
Prompted or scripted speech recorded in controlled conditions with matching transcripts.
Sung audio
Sung vocal recordings with performer consent, for music and voice modelling.
Film & professional video
Professionally shot film, broadcast, and catalog footage available for AI licensing.
Creator & UGC video
Creator and user-generated video sourced directly from the people who filmed it, with consent.
Egocentric video
First-person video from head-mounted and wearable cameras, capturing real tasks from the doer’s viewpoint.
Motion capture
Marker and markerless motion capture from consented performers, delivered rig-ready.
Sensor & IMU
Inertial and device-sensor data streams paired with activity labels.
Agentic trajectories
Step-by-step records of humans completing digital or physical tasks, for training agents.
LiDAR & 3D point clouds
LiDAR data records the distance from a sensor to surfaces as a point cloud, often paired with sensor pose, timestamps, intensity values, maps, images, or semantic labels.
BUILT AROUND YOUR BRIEF
Need a different collection?
Describe the task, participants, recording conditions, volume, and timeline. We will assess what can be sourced or commissioned for your requirements.
Start with a sample.
Tell us the collection and task you want to evaluate. We’ll confirm the available sample and the appropriate review scope.
Request a sample