Agents call participants on schedule
Voice agents place calls on the protocol's cadence, in the participant's home, with no app to install and no site visit required.
For research
A managed voice AI platform for decentralized trials. We run the calls, the collection and the longitudinal analysis in each participant's natural environment, so your team keeps its focus on the science.
What it does
Agents reach participants on schedule, capture structured and naturalistic data, and turn recurring calls into longitudinal signal on voice and verbal patterns.
Voice agents place calls on the protocol's cadence, in the participant's home, with no app to install and no site visit required.
Each call gathers the structured items your protocol requires alongside naturalistic conversation, capturing both prompted responses and spontaneous speech.
Voice signals and verbal patterns are analysed over time, building a per-participant trajectory rather than a series of isolated snapshots.
Why phone-based
Voice-first acquisition removes most of the friction that limits digital endpoint capture in older or cognitively vulnerable populations — no wearables, no apps, no participant burden.
Daily mood signal between site visits with weak-signal alerting.
Longitudinal cognitive markers without repeated formal testing.
Speech-based features and adherence reporting across the home setting.
Daily affect and engagement signals between clinical contacts.
Patient-reported symptom and impact tracking via natural conversation.
Parallel signal capture from family informants where appropriate.
Operated for you
Managed means managed — the day-to-day of running calls and data flow sits with us, not your coordinators.
Call cadence, retries and windows are managed for you, following the protocol without adding coordinator workload.
Calls run only within the consent you have on file, and the agent identifies as AI on every contact.
Study teams see adherence, call status, signal trends and flags in one place, in near real time.
Structured per-participant data is exported to your systems on the schedule your team needs.
Longitudinal analysis
The value of phone-based monitoring is resolution over time. The analysis is built to read each participant's trajectory against their own history.
The optional marker layers are the same engines behind CogPath and VGDS, and a dedicated biomarkers extraction pipeline can be built for custom endpoints.
Compliance
De-identification, audit lineage and a versioned pipeline are built into how the platform is operated, not added at the end.
Works with your protocol
Outputs map to your eCRF and EDC and export in open formats, so the platform fits the tools your study already runs on.
Structured outputs map to your eCRF and EDC, with standard CDISC mappings available on request.
Per-participant time-series data is delivered as JSON, CSV or Parquet, with a versioned data dictionary attached.
Need the platform to be yours to own and operate rather than run as a service? See bespoke Voice AI for decentralized trials.
Share your protocol, endpoints and population, and we will map a managed voice AI deployment that fits — calls, collection and longitudinal analysis, operated for you.