MP3 per chapter
Plays on any phone, car stereo or old MP3 player. A good choice when a store or distributor wants separate chapter files.
Word to audiobook converter
Skip the export step. Upload the manuscript you edit in Microsoft Word and the Word to audiobook converter turns each Heading 1 or Heading 2 into a chapter, gives the narrator and every speaking character a voice, and lets you download MP3 or M4B files.
DOCX files · also EPUB, PDF, TXT, MD · up to 30 MB
Read in your browser. Nothing is uploaded until you generate
A Word document to audiobook conversion needs no plugin or extra software. The first step only matters for older files.
Older .doc files will not open. In Word, choose File > Save As and pick Word Document (.docx). Pages, LibreOffice and Scrivener can export the same format.
Drag it onto the uploader above. The file is read in your browser, up to 30 MB, and nothing is sent anywhere until you generate.
Each heading appears as a chapter. Characters found in the dialogue are listed with suggested voices that you can swap before generating.
Generate a chapter, listen, then do the rest. Save one MP3 per chapter, a ZIP of the whole book or an M4B with named chapter markers.
A DOCX is a structured file, so most of what you see on screen has a clear meaning in the audio. This is how each part is treated when you convert Word to audiobook.
| In your Word file | In the audiobook |
|---|---|
| Heading 1 or Heading 2 style | Starts a new chapter named with the heading text |
| A typed line such as “Chapter 4” or “Part Two” | Also starts a chapter, even without a heading style |
| Text before the first heading | Kept as an “Opening” section if it runs past about 120 words |
| Body paragraphs | Narration, with quoted dialogue passed to each speaker |
| Headers, footers, footnotes and comments | Not read, because Word stores them outside the main text |
| Tracked changes still pending | Insertions are read and deletions are not, so resolve them first |
| Title in File > Info > Properties | Becomes the book title shown in audiobook players |
| Images, charts and text boxes as pictures | Skipped; only words are narrated |
Writing in Google Docs instead? Download a DOCX first, as described on the Google Docs to audiobook page.
[say with hyped, high energy like a YouTube intro]
Stop scrolling. In the next sixty seconds, I'm going to show you the trick that doubled my channel.
Your manuscript stays a manuscript. In a DOCX to audiobook project, delivery notes are added to lines in the studio after upload, so you never have to fill the Word file with stage directions that would later end up in your ebook.
Open any line and type a note such as [whisper], [furious] or [like a tired detective at 3 a.m.] at the start. Sound tags such as [sigh] or [laugh] can sit inside a line on the Standard, Lite and Expressive models. The AI voice emotion tags page lists what works.
Check your headings in the Navigation Pane. In Word, open View and tick Navigation Pane. Every chapter should be listed there. If one is missing, click its heading and apply Heading 1 from the Styles gallery. Headings longer than about 100 characters are read as body text, so keep chapter titles short.
Put dialogue in double quotation marks. Speakers are found from text inside double quotes (straight or curly) plus a tag such as “said Nora” or “Nora asked”. Books that use single quotes for speech, common in British publishing, should switch for the audio copy. Work on a duplicate and check each change, since a closing single quote is the same character as an apostrophe.
Accept or reject tracked changes. Open the Review tab and choose Accept All Changes (or work through them one by one). Pending insertions would otherwise be narrated alongside the final text, and the audio should match the edition readers buy.
Spell out what a voice might stumble on. Symbols, abbreviations and invented names are where narration slips. Write “Number 3” instead of “No. 3”, and keep a short list of tricky names to fix by ear. The audiobook pronunciation guide shows respellings that voices follow.
The Read Aloud button on Word’s Review tab speaks the document in one voice while you follow along. It is handy for catching typos, but it gives you no file, no chapters and no character voices. A Word to audiobook export gives you MP3 or M4B files you can keep, share with beta listeners or publish.
Every DOCX to MP3 or M4B export comes from the same generated audio, so you can download more than one.
Plays on any phone, car stereo or old MP3 player. A good choice when a store or distributor wants separate chapter files.
One file with named chapter markers and title tags. Apple Books, BookPlayer and Audiobookshelf open it with bookmarks.
Every chapter file plus a chapters index in one download, handy for archiving a finished edition or handing it to a distributor.
Importing, finding the chapters and casting the voices are free for any Word to audiobook project. Generating audio uses credits charged per 1,000 characters. Standard voices work out at roughly $9 to $12 per finished hour, so an 80,000 word manuscript (about nine hours) lands around $80 to $110. Sign up with Google and you get 100 gift credits (valid for 7 days, about 12 minutes on Standard; usage varies by text and voice tier) for a short sample, and plans are on the pricing page.
If the manuscript is your own, you own the audio you make from it. Where you can sell it depends on each store’s rules for AI narration, which change; ACX, for one, turns away author-uploaded AI narration at the time of writing. The manuscript to audiobook guide walks through a full indie release, and the AI audiobook generator home page has more sample casts.
Not as it is. Only .docx files open in the converter. Open the .doc in Word, choose File > Save As and pick Word Document (.docx), then upload the new copy. Your text and heading styles come across unchanged, and the Word to audiobook conversion works as normal.
Heading 1 or Heading 2 is the most reliable way, but typed lines that begin with Chapter, Part or Book followed by a number also start a chapter. If neither is present, long text is split into parts of about 4,000 words, which you can still generate one by one.
No. Footnotes, endnotes, comments, headers and footers are stored outside the main body of a DOCX, so they stay out of the audio. If a footnote matters to the story, move its text into the body before uploading, as a sentence the narrator can read naturally.
Upload the DOCX, check the chapters and voices, generate the audio and choose the MP3 export. You get one MP3 per chapter, or a ZIP with every chapter file and an index. The same audio can also be saved as an M4B audiobook.
Word can speak a document through Read Aloud, but at the time of writing it does not save that speech as an audio file. To get files, upload the DOCX here and export MP3 or M4B with a separate voice for each character.
Yes, but only regenerate what changed. Edit the affected line in the studio and regenerate that line alone; the rest of the chapter stays as it was. For bigger rewrites, upload the revised DOCX as a new project and reuse the same cast.

Upload a DOCX and try Word to audiobook free on a short sample.