Audiobook Chapter Splitter

Why one giant file is a problem

A 12-hour audiobook as a single MP3 is hostile to every player on earth. Lose your place and you're scrubbing through a progress bar the width of a continent; want to re-listen to chapter 14 and you're guessing at timestamps; sync it to a car stereo or an old MP3 player and bookmarks simply don't exist. Splitting into per-chapter files fixes all of it: each file is a chapter, players remember your position per file, and skipping ahead or back becomes one tap. It's also how audiobooks were meant to be distributed — Audible and Apple Books both deliver chaptered files under the hood, they just hide the plumbing from you. This tool gives you that same plumbing for any long recording: audiobooks, lecture series, conference recordings, or that 8-hour podcast archive you downloaded as one file.

M4B chapters: the exact method

Professionally produced M4B audiobooks usually carry embedded chapter markers (the Nero "chpl" format): precise timestamps plus real chapter titles, written by the publisher. When this tool finds those markers, it uses them verbatim — no guessing, no silence analysis, and your exported MP3s get the publisher's actual chapter names. That is why the "use embedded chapters" option is on by default and checked first. Note the honest limitation: some M4B files store chapters as a separate text track instead of Nero markers, and a few use proprietary layouts. Those variants aren't readable with client-side code alone, so the tool falls back to silence detection for them — which is still a solid starting point you can refine by hand.

Tuning silence detection

When there are no embedded chapters, the tool analyzes audio energy in 50-millisecond windows and flags every silence longer than your "min silence" setting that falls below the "silence threshold". Three knobs control the result. The silence threshold (default -35 dB) decides what counts as quiet: raise it (toward -20 dB) if the narrator's pauses are soft or the recording has room tone; lower it (toward -50 dB) if music interludes are being mistaken for chapter breaks. Min silence (default 1.2 s) filters out dramatic pauses mid-sentence — chapter breaks in audiobooks are usually 2–4 seconds of dead air. Min chapter length (default 4 min) merges splits that land too close together, which kills most false positives from intro music. Start with the defaults, look at the proposed splits on the waveform, and only then touch the knobs — two passes beat ten minutes of fiddling.

Organizing the exported chapters

The ZIP contains files named like 01 - The Beginning.mp3, numbered so every player and file manager sorts them correctly (the zero-padded numbers matter — "10" sorts before "2" without them). Filenames are sanitized so they work on Windows, macOS and Linux. A few practical tips: pick 96 kbps for spoken word — it's transparent for voice and roughly halves the size versus 128 kbps; keep the chapter titles short since some car stereos truncate long filenames; and if you use a phone player with its own bookmarking (like Smart Audiobook Player or Bound), the per-chapter files give it clean resume points for free. Your original file is never modified — splitting only reads from it, so experiment without fear.