One-Shot Summary of an Audio File via Cochl.Sense
Send in a recording and get back a short JSON summary of the whole file: whether there is speech, what language, what kind of place it sounds like, what happened, and keywords. Runs on Cochl's cloud with your own Cochl project key, so it uses 0 CNAPS credits.
🔑 Requires a Cochl project key
What it does
Audio Insights listens to a whole audio file and describes it in one structured summary. It combines Cochl's sound event detection with its speech analysis, so it tells you both what was said and what else was going on — a meeting with a phone ringing, a street interview with sirens behind it, a quiet room with typing.
It gives a summary, not a timeline. There are no timestamps in the output. When you need to know when something happened, use Sound Events (Cochl) or Speech Analysis (Cochl) instead.
Problem it solves
Triage a pile of recordings | Know what each file is about without listening to it |
Auto-tagging | Keywords and a sound-environment label for search or a media library |
Language routing | A detected language code to pick the right next step (translation, TTS voice) |
Context for an LLM | Feed the JSON to a language model to write a caption, title or report |
Input/Output
- Input: Sound — one audio file, up to 1 hour. MP3, WAV, FLAC and OGG are sent as-is; other formats are converted first. Starting from a video? Use Video Audio Extract first
- Output: Text — a JSON summary with these fields:
- contains_speech — whether the file has speech
- detected_language — a language code such as "en" (not a language name)
- primary_sound_environment — a short label such as "office/meeting"
- situation_summary — what is going on in the recording
- notable_events — sounds worth calling out
- speech_content_summary — what was said
- keywords
- Settings: none
- Requires: A Cochl project key, added under your avatar → Workspace Settings → Integrations
How to Use
- Add your Cochl project key under Workspace Settings → Integrations.
- Upload the recording with Sound Loader. For a video, use Video Loader → Video Audio Extract.
- Connect the audio to Audio Insights (Cochl) and the output to Text Viewer.
- Run and read the summary. To turn it into prose, connect the output to a language model.
Example Workflow
Flow: Sound Loader → Audio Insights (Cochl) → Text Viewer
Video source: Video Loader → Video Audio Extract → Audio Insights (Cochl) → Claude or Gemini ("Write a one-line caption and five hashtags from this summary") → Text Viewer
Choosing an audio understanding model
You want | Use |
A quick overview of the whole file — speech, language, setting, events | Audio Insights (Cochl) |
Non-speech sounds with start and end times | Sound Events (Cochl) |
A transcript split by speaker, with timestamps | Speech Analysis (Cochl) |
Plain transcript text via Google | Transcribe (Gemini) |
Timed transcripts on CNAPS servers, no outside key | Speech Recognition |
Performance
- Execution: Cochl cloud API; no local GPU
- Length: up to 1 hour of audio per run
- Billing: 0 CNAPS credits. Cochl bills your project, and every Audio Insights run counts as Sound Event Detection minutes
- Not for: anything that needs timing — this output has no timestamps
Technical Details
Service | Cochl.Sense Cloud API — Audio Insights |
Developer | Cochl |
How it works | Derived from Cochl sound event detection plus speech analysis |
Output format | JSON text, no timestamps |
Runtime | Cochl cloud, with your Cochl project key |
Compliance & Provenance
Provider | Cochl |
Provider type | External API |
License / service terms | Cochl service terms and the terms of your Cochl project |
Output provenance | A summary derived from the audio; see CNAPS AI Policy for attribution and metadata |
For more on how we classify models and mark outputs, see our AI Policy.
Model Source
- Cochl.Sense
- Cloud API docs: Cochl.Sense Cloud API — Getting started