Skip to content

Try Resonance 2 early.

Tell us a little about yourself and what you’re building.

A brief description of your use case.

We’ll use these details to review your application and contact you about access. Privacy policy

Get started free

Explore oruk

Qualitative research

Find interview passages worth a closer listen

Use transcripts and timed acoustic labels to search and compare recorded interviews. Researchers review the context and decide how to interpret each passage.

For teams building this workflow into their own product. Start with prerecorded English audio; Oruk returns structured analysis through the API.

Try your audio Build with the APIEstimate cost

No account needed to try the public demo.

Illustrative output

Suggested annotations alongside the transcript

I was relieved when the process ended, but I am still uncertain about what comes next.

relieved 0.79worried 0.58sincere 0.67

Example labels for this interface concept. Analyze your recording to inspect actual model output.

Unified analysis request
curl https://speech-api.oruk.ai/v1/audio/analysis \
  -H "Authorization: Bearer $ORUK_API_KEY" \
  -H "X-Request-ID: $(uuidgen)" \
  -F "model=oruk-resonance" \
  -F "file=@recording.wav"

Example implementation

Where Oruk fits in your workflow

Your input
Consented, prerecorded English interviews and the study’s codebook. Keep participant identifiers in your research system.
Oruk returns
Timed transcript segments with suggested acoustic labels and model metadata that can accompany your coded excerpts.
Your application
Add candidate tags to your coding workspace, link them to the original recording, and let researchers accept, revise, or reject each suggestion.

Evaluate the result in your pilot

Compare coding time and reviewer agreement on a held-out interview set. Record overrides and examine where accents, context, or recording quality change the annotations.

How it works

01

Define the codebook

Choose which API labels are relevant and document what they do and do not mean in the study.

02

Batch prerecorded files

Analyze recorded interviews and retain the model version and request ID with each result.

03

Review and adjudicate

Researchers inspect passages, revise suggested labels, and document how model outputs informed their coding.

What you can build

  • Find candidate passages for close reading
  • Compare aggregate expression patterns across interview sets
  • Attach reproducible acoustic features to coded excerpts
  • Prioritize human review without replacing it

Deployment safeguards

  • Obtain research consent for automated speech analysis
  • Do not treat model labels as participants’ self-reported feelings
  • Evaluate cultural, linguistic, and recording-condition differences
  • Document model version, thresholds, overrides, and missing data
Responsible-use guidance

Related resources

AI call analytics Contact center analytics Voice call sentiment analysis Voice of customer analytics Sales call analysis Meeting speech understanding Speech analytics API Speech emotion recognition API API capabilities and scope API documentation Build a speaker-turn review workflow Stream transcription and phrase emotion Pricing