How to split audio into chapters without uploading
- Step 1Read your chapter start times — Open the audiobook or lecture in a player and note each chapter's start time (mm:ss or hh:mm:ss). The splitter can't read embedded chapter tags, so you'll enter these yourself. A chapter list from the publisher works too.
- Step 2Open the splitter and drop the file — Go to the audio splitter and drop the audiobook or lecture (one file). It loads into memory in your browser and is never uploaded — the source stays on your device.
- Step 3Set Split mode to Manual markers — Change the Split mode dropdown to Manual markers. The marker list and an Add marker button appear. (Every N seconds and On silence modes exist too, but chapters have specific starts, so markers are the right choice.)
- Step 4Convert each chapter start to seconds — Markers are in seconds. For hh:mm:ss, compute hours×3600 + minutes×60 + seconds: 1:12:30 → 3600 + 720 + 30 =
4350. The field accepts one decimal place for fine boundaries. - Step 5Add a marker per chapter start — Click Add marker for each chapter boundary (the start of chapter 2, chapter 3, …). You don't mark chapter 1's start (0) — the file already begins there. N chapters need N−1 markers. Order doesn't matter; they're sorted.
- Step 6Run and download each chapter — Start processing. Each chapter is re-encoded and listed with its size. Click Download all to save every chapter as its own file, or Download per chapter. Files are named
<name>-part-N; rename them to chapter titles afterward.
hh:mm:ss to seconds for chapter starts
Markers are entered in seconds. Audiobooks and lectures often exceed an hour, so convert hh:mm:ss with hours×3600 + minutes×60 + seconds.
| Chapter start | Seconds | Chapter start | Seconds |
|---|---|---|---|
| 0:08:00 | 480 | 1:12:30 | 4350 |
| 0:21:45 | 1305 | 1:48:10 | 6490 |
| 0:42:00 | 2520 | 2:05:00 | 7500 |
| 0:55:30 | 3330 | 2:40:15 | 9615 |
| 1:00:00 | 3600 | 3:00:00 | 10800 |
Chapters from markers
Chapter 1 starts at 0 (no marker needed). Add a marker at each subsequent chapter's start. Chapters = markers + 1.
| Chapters wanted | Markers to add | Markers = chapter 2..N starts |
|---|---|---|
| 3 chapters | 2 | start of ch2, start of ch3 |
| 6 chapters | 5 | starts of ch2 through ch6 |
| 10 chapters | 9 | starts of ch2 through ch10 |
| N chapters | N − 1 | every chapter start except chapter 1 |
What the chapter split does and does not do
Grounded in the tool. Where a capability isn't present, a sibling tool or manual step is noted.
| Capability | Supported? | Detail / alternative |
|---|---|---|
| Read embedded chapter tags | No | Enter chapter starts manually as markers |
| No upload (fully local) | Yes | FFmpeg 8.1 WebAssembly in-browser |
| Keep source format | Yes | No format selector; output matches input |
| Lossless / stream-copy cut | No | Re-encoded; use audio-trimmer for lossless |
| Name chapters by title | No | Files are -part-N; rename after download |
Cookbook
Chaptering recipes for audiobooks and lectures. Chapter starts are entered in seconds (convert hh:mm:ss). Chapters download as separate files and keep the source format; the file never uploads.
Split a 6-chapter audiobook by published chapter starts
The publisher lists chapter start times. Convert each (except chapter 1) to seconds and add as markers.
Input: book.m4a (3:05:00) Chapter starts (ch2..ch6): 0:42:00, 1:12:30, 1:48:10, 2:05:00, 2:40:15 Mode: Manual markers Add markers (s): 2520, 4350, 6490, 7500, 9615 Result: 6 chapters (markers + 1) book-part-1 ... book-part-6 (each .m4a)
Chapter a lecture you timestamped while listening
You noted section starts during playback. Enter them as markers; each section becomes a chapter file.
Lecture sections start at: 8:00, 21:45, 42:00, 55:30 Mode: Manual markers Add markers (s): 480, 1305, 2520, 3330 Result: 5 chapters (intro + 4 sections) (file never leaves the browser)
Convert an M4B audiobook first, then chapter it
If your player wants MP3, convert the M4B's audio to MP3, then chapter the MP3 with the same markers.
1) m4a-to-mp3 (/audio-tools/m4a-to-mp3) book.m4b audio -> book.mp3 2) splitter, Manual markers at chapter starts -> book-part-1.mp3 ... book-part-N.mp3
Fine boundary so a chapter doesn't clip the last word
A chapter starts mid-sentence at 42:00.0 but the previous chapter's last word ends at 41:59.6. Nudge the marker.
Default marker: 2520 (42:00.0) clips last word Adjusted marker: 2519.5 (41:59.5) -> previous chapter keeps its full sentence
Rejoin two chapters that were split too finely
Chapters 4 and 5 are really one part; merge them after splitting.
splitter -> 6 chapters audio-merger (/audio-tools/audio-merger) -> drop book-part-4 + book-part-5 in order -> one combined chapter file
Edge cases and what actually happens
Audiobook has embedded chapter markers
Not readThe splitter does not parse ID3 chapter tags, M4B/Nero chapters, or Podlove chapters. Read the chapter start times from your player or the publisher's chapter list and enter them as manual markers (in seconds). The tool cuts exactly at the points you provide.
Marking chapter 1's start at 0
IgnoredA marker at 0 is dropped because the file already starts there. You only need markers for chapters 2 through N. Adding a 0 marker won't create a duplicate or empty chapter — it's simply filtered out.
Entering a chapter time as hh:mm:ss
Invalid inputThe marker field takes seconds, not clock time — 1:12:30 won't parse. Convert with hours×3600 + minutes×60 + seconds (1:12:30 = 4350). The conversion table above lists common values for hour-plus chapters.
A chapter cuts a word or two early/late
Use a decimal markerWhole-second markers can clip the tail of the previous chapter's sentence. Nudge the boundary with a decimal (e.g. 2519.5) — the field accepts 0.1 s steps — so each chapter keeps its full audio.
Chapters come out re-encoded
Re-encodedEach chapter is re-encoded (M4A via AAC, MP3 via libmp3lame), not stream-copied — one lossy generation for compressed sources. For listening this is inaudible at audiobook bitrates. For a byte-exact lossless single cut, use the audio trimmer's lossless mode.
Wanting chapters in a different format than the source
Convert firstNo format selector means output matches input. To get MP3 chapters from an M4B/M4A audiobook, convert with m4a-to-mp3 first, then split. The chapters then come out as MP3.
Chapters named by index, not title
Rename afterOutput files are <name>-part-1, -part-2, etc. The tool has no title field, so chapter titles aren't written into filenames or tags. Rename the files after download to match the chapter list.
Audiobook longer than your tier's duration limit
RejectedPro caps a single file at 120 minutes — many audiobooks are longer. A 3-hour title is rejected on Pro; use Pro-media or Developer (unlimited duration, 100 GB), or split the source into two halves first with the audio trimmer.
Free tier user tries to chapter a book
Tier blockedThe splitter is Pro-tier and won't run on Free. Pro suits shorter books and lectures (up to 120 minutes); full-length audiobooks generally need Pro-media for the unlimited duration.
Worried the audiobook is uploaded
Stays localIt isn't. FFmpeg 8.1 runs in your browser via WebAssembly and the file is processed entirely on your device — nothing is transmitted. This is what makes chaptering licensed or unpublished content safe here.
Frequently asked questions
Can it use my audiobook's built-in chapter markers?
No — the splitter doesn't read embedded chapter tags (ID3, M4B/Nero, or Podlove). Read each chapter's start time from your player or the publisher's chapter list, convert to seconds, and enter them as manual markers. The tool then cuts exactly at those points.
How do I split an audiobook into chapters?
Set Split mode to Manual markers, then click Add marker for each chapter start (in seconds — 0:42:00 is 2520). You don't mark chapter 1 (it starts at 0). N chapters need N−1 markers; the result is one file per chapter.
Is the audiobook uploaded anywhere?
No. FFmpeg 8.1 runs in your browser through WebAssembly, so the file is split entirely on your device and nothing is transmitted. That's what makes chaptering purchased or unreleased content safe.
How do I convert hh:mm:ss to the seconds the marker field wants?
Multiply hours by 3600, minutes by 60, and add the seconds. 1:12:30 = 3600 + 720 + 30 = 4350. The conversion table above covers common hour-plus chapter starts, and the field accepts one decimal for fine boundaries.
Why doesn't '1:12:30' work in the marker field?
It's a numeric seconds input, not a clock field. Convert to seconds first (1:12:30 = 4350). Entering clock notation won't parse.
Are the chapters lossless?
Each chapter is re-encoded (M4A via AAC, MP3 via libmp3lame), so compressed sources get one lossy generation — inaudible at audiobook bitrates. For a byte-exact lossless cut, use the audio trimmer's stream-copy option, one section at a time.
What format are the chapter files?
The same as the source. There's no format selector, so an M4A audiobook produces M4A chapters and an MP3 produces MP3 chapters. Convert with m4a-to-mp3 first if your player needs MP3.
Can I name the chapters by title?
Not in the tool — chapters are named <name>-part-N by index. The splitter has no title field, so rename the downloaded files to your chapter titles in your file manager.
Do the chapters download as a ZIP?
No — each chapter is a separate file. Download all saves them individually in one click; per-row Download links grab a single chapter. There's no ZIP option.
My chapter boundary clips the last word — how do I fix it?
Nudge the marker by a fraction of a second. The field accepts 0.1 s steps, so change 2520 to 2519.5 to keep the previous chapter's final word. Re-run after adjusting.
Can I chapter a full-length (3-hour) audiobook?
Not on Pro, which caps a single file at 120 minutes. Use Pro-media or Developer (unlimited duration, 100 GB file limit), or split the audiobook into two halves first with the audio trimmer and chapter each half.
Can I merge two chapters back together if I split too finely?
Yes — drop the chapter files into the audio merger in order and it concatenates them end-to-end into one file. Useful when two short sections are really one chapter.
Privacy first
Every JAD Audio tool runs entirely in your browser via FFmpeg (WebAssembly) and RNNoise. Your audio files never leave your device — verified by zero outbound network requests during processing.