# Call analytics

- Page: https://scanovich.ai/projects/call-analytics/
- Language: en
- Updated: 2026-07-17

- Status: prototype
- Code: public code
- Repository: https://github.com/FUYOH666/Scanovich.ai-audio-call
- Research line: [Private, local AI](https://scanovich.ai/research/private-local-ai.md)

Call recordings become transcripts and structured quality reviews on infrastructure the owner controls.

Speech recognition, clean-up, classification and review notes, with audio and transcripts kept on the owner's disks.

## What works today

- One pipeline behind three entry points: web, HTTP and command line.
- A reference configuration tuned for one class of telephone recordings.

## Why it is in the lab

Review teams usually listen to a small sample of calls. The question is how much of the review can be prepared locally, and what has to be retuned for each language and call type.

## Evidence

- **Reference stack** — A working pipeline behind a web interface, an HTTP interface and a command line. Measured 2026-07-17. Source: https://github.com/FUYOH666/Scanovich.ai-audio-call

## How to engage

A pilot on your recordings — tuned to your language, audio quality and review criteria.

[Contact](https://scanovich.ai/contact/) · private@scanovich.ai