Skip to main content
Cognitive analysis is one of Dolva’s two core analysis types. When you send an audio recording to the /v1/analyze/cognitive endpoint, Dolva’s models examine the acoustic and speech-pattern properties of the recording to surface signals about the speaker’s cognitive state — for example, how much mental effort they appear to be exerting, how clear their thinking sounds, and related markers.

​
What It Measures

Cognitive signals are extracted from the natural properties of speech — things like pacing, hesitation patterns, vocal cadence, and articulation. These acoustic features correlate with well-studied markers of cognitive engagement and load. The analysis is fully automatic: you supply the audio file and Dolva handles all signal extraction. No transcription is required, and the models work directly on the audio waveform.

​
Common Use Cases

  • Mental wellness monitoring — Track cognitive changes over time for individuals or cohorts
  • Conversation intelligence — Assess cognitive engagement in sales calls, interviews, or coaching sessions
  • Clinical research — Supplement clinical assessments with objective acoustic signal data
  • Accessibility tooling — Detect cognitive fatigue in real-time applications

​
How to Request Cognitive Analysis

Send a multipart/form-data POST request to /v1/analyze/cognitive with your audio file:
curl
curl -X POST https://api.dolva.ai/v1/analyze/cognitive \
  -H "Authorization: Bearer dv-xxxxxxxx" \
  -F "audio=@conversation.wav"
The response is a JSON object containing the cognitive signal data extracted from your audio.

​
Relationship to Emotion Analysis

Cognitive and emotion analyses are independent — you can run one or both on the same recording. While they analyze the same audio file, they target different signal dimensions: cognitive analysis focuses on processing patterns and mental load, while emotion analysis targets affective states and valence.

View the API Reference

Full request/response details for POST /v1/analyze/cognitive