DAILY BRIEFING / LOCAL AUDIO

Your briefing, in your voice.

Download a briefing PDF from the dashboard, then upload it to F5 Cyber Audio on your computer. Chatterbox turns the narration into a WAV file. There is no subscription; audio length follows the script.

1. Get the local app

Install Python 3.12. Create a folder called F5Cyber-Audio and save these files inside it:

Mac installation

Also save Install and Start.command and Start Audio.command in that folder.

Open Terminal, type bash followed by a space, drag Install and Start.command into the window, and press Return. The installer creates an isolated environment, installs Chatterbox, and opens the app. For later launches, repeat with Start Audio.command.

Apple silicon uses its GPU when available. Older Macs use the CPU and may take longer.

Windows installation

Open Terminal in the folder and run these lines in order:

py -3.12 -m venv .venv
.venv\Scripts\python -m pip install -r requirements.txt
.venv\Scripts\python app.py

Use the last line to open the app again. NVIDIA acceleration depends on a compatible PyTorch installation; CPU generation is slower.

Linux installation
python3.12 -m venv .venv
.venv/bin/python -m pip install -r requirements.txt
.venv/bin/python app.py

2. Download today’s briefing

On the dashboard, choose Create audio briefing (PDF). It uses today’s top five public stories across all categories, ranked by known exploitation, then EPSS, then CVSS. Dashboard filters do not change this edition. If no stories are dated today (UTC), wait for a feed refresh.

3. Upload, review, listen

  1. With the local app running, open F5 Cyber Audio. If it cannot connect, launch the app first.
  2. Upload the PDF. Its included narration script excludes page numbers and the source appendix. Review and edit it before generation.
  3. Optionally upload 10–20 seconds of clear speech in your own voice. A quiet room and an unprocessed WAV recording work best. Without a sample, the app uses its default voice.
  4. Confirm voice permission and choose Generate audio. Listen, then download the WAV. Keep the Terminal window open while it works.

Review names, acronyms, and factual accuracy before publishing. The briefing uses source excerpts; it does not add independent reporting. Chatterbox adds a watermark to generated audio.

Help & privacy

Generic PDFs need selectable text; scanned or encrypted PDFs are not supported. For long documents, split the narration into sections under 25,000 characters. Your generated files are in the app’s output folder. Close Terminal to stop the app.

F5 Cyber Audio runs at a local address and never creates a public sharing link. Only reproduce your own voice or one you have permission to use.

Built with Resemble AI’s MIT-licensed Chatterbox. Installation has been tested on Apple silicon; other platforms may need additional dependency setup.