Build a local call-review panel with Python
An audio analysis response is easier to inspect when you can hear the passage beside its transcript. This tutorial builds a local review page that lets you replay a speaker turn, search the words or annotations, and read vocal-expression scores separately from what was said. We will start with an actual saved response and a licensed public recording. That path needs Python and a browser, but no…
This guide demonstrates how to create a local call review panel using Python. The process begins with an actual saved response and a licensed public recording. You will need Python 3.10 or later, curl, and a current browser. The tutorial provides a complete renderer using Python's standard library, generating an HTML file with an audio player and plain JavaScript controls.
To run the saved example, download the necessary files and execute the call-review.py command, passing the response.json, recording.wav, output review.html, and attribution.txt files. Open the generated review.html in your browser to review the call.
To analyze your own recording, create an English recording with permission to process, up to 30 MB and 60 minutes. Set the ORUK_API_KEY in your environment and run the analyze-file.py command, which uses the Oruk SDK to analyze the file and generate a JSON response. Then, use the call-review.py command to generate the review.html file. The report is built around a small data model, displaying segment start and end times, speaker label, words, emotion scores, and style scores.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
