Try Pro FREE for 2 weeksGet Started
Start for free

cochl-sense-audio-insights — LLM AI Model

Cochl.Sense listens to a whole recording and writes up what it contains: the sound environment, a short description of the situation, notable sound events, whether there is speech, its language and a summary of what was said, plus keywords. Runs on Cochl's side against your own project key.

Run cochl-sense-audio-insightsBuild with it in Studio

What it does

Audio Insights (Cochl) summarises a recording of up to an hour as a whole. It reports the main sound environment as a short label such as office/meeting, a short description of the situation, notable sound events, whether anyone speaks and the language code (en, ko and so on), a summary of what was said and a set of keywords.

Cochl builds the summary from its sound event detection and speech analysis of the same file, so it is the quickest way to know what a recording is about before deciding what to cut or transcribe.

When to use it

  • Triage of long recordings: what is in each file before anyone listens to it
  • Keywords and a one-paragraph description for cataloguing audio or video archives
  • Checking whether a clip contains speech, and in which language, before transcribing it

Details

  • CategoryLLM
  • Year2026
  • LicenseCustom (Cochl Terms of Service)

Compliance & Provenance

  • ProviderCochl · Specialized
  • EU AI Act RiskMinimal Risk
  • Art. 50 TransparencyNot applicable

Inputs & Outputs

SoundInputsound

Recording to summarise (up to one hour)

SummaryOutputstring

Summary of the recording as JSON or as readable text, per Output format

Parameters

cochl_output_formatstringdefault JSONoptions: JSON, Text

Output format: JSON for other nodes to read, Text to read yourself

How the summary is written. 'JSON' (default): the object the model description lists. 'Text': plain 'Label: value' lines - Setting, Summary, Sounds (Cochl tags, underscores as spaces), Speech (yes/no with the language code), What was said, Keywords - then any other field Cochl returns, under its own key. Empty fields are left out; if every field is empty the output is 'Cochl returned no insights.' The labels are English whatever the recording's language.

Limitations

  • Up to one hour of audio per run. mp3, wav, flac and ogg are sent as they are; other formats are converted first.
  • Cochl runs sound event detection to produce the summary, so each run is billed as Sound Event Detection minutes on your Cochl project.
  • The output describes the whole file, as JSON or (Output format: Text) as readable labelled lines; it has no timestamps. Use Sound Events or Speech Analysis for timed results.
  • Runs on your own Cochl project key. Without one configured in the workspace the node cannot run, and usage is billed by Cochl to that project.

Tags

  • audio-understanding
  • summarization
  • sound-event-detection
  • audio

Alternatives in Audio Understanding

  • Sound Events (Cochl)

    Cochl.Sense sound event detection: finds 100+ kinds of sound (glass breaking, siren, dog bark, applause...) and when they happen. Needs Cochl key.

  • Speech Analysis (Cochl)

    Cochl.Sense speech analysis: speaker-separated transcript in 100 languages, naming voices registered in your Cochl dashboard. Needs Cochl key.

  • Speech Recognition

    Recognize speech in a video or audio file and convert it to text.

  • Speech Recognition (Clips)

    Recognize speech in each clip of a video list and convert it to timed transcripts.

  • Transcribe (Gemini)

    Google Gemini 3.5 Transcribe: speech to text in 85+ languages, with optional speaker labels and custom vocabulary. Needs API key.

Resources

All AI models