Lyrics guidelines
Lyrics must match the audio word for word and follow store formatting rules — spacing, capitalisation, punctuation, slang, numbers, spoken parts — plus the exact timestamp format for synced lyrics and the mistakes that break it.
Part of the ANS Music Audio Content Guidelines. Lyrics let listeners follow along during playback — and stores display them exactly as you submit them. Lyrics must match the audio verbatim and follow the rules below so they’re consistent across every platform.
- Rule one Transcribe exactly what is sung — nothing more, nothing less
- Where to add them Per track, in Backstage, during release creation
- Synced lyrics Timestamp each line:
[mm:ss.mmm] Text
Spacing and structure#
- Lines within a single verse or chorus are single-spaced.
- Section or stanza breaks are shown with one blank line.
- Don’t include text that isn’t audibly present — no section headers, no vocalist names. No “Intro”, “Verse”, “Chorus”, “Hook”, “Outro”.
- Repeated lines and choruses must be written out each time. Never write “Chorus 2x” or “(x2)”.
Capitalisation#
- Begin each line with a capital letter.
- Proper nouns — titles of films, TV shows, songs — are title-cased.
- All caps may be used to mark an exclamation.
Punctuation and formatting#
- No full stops or commas at the end of a line or section.
- Question marks and exclamation marks are included at the end of a sentence where appropriate.
- No italics or bold.
- Background singers and hushed vocals may be transcribed in parentheses if they contribute to the content of the song.
- If a song ends with repetitions that fade out, transcribe the first four repetitions and end the final line with an ellipsis (…).
Slang and censored language#
- When providing lyrics for the clean version of a song that also has an explicit version, use asterisks to represent the censored audio.
- Use the contractions and slang that match the audio: ain’t, ’bout, ’til, ’cause, fallin’, walkin’.
Numbers#
- Numbers are written out as words unless they’re commonly written in numerals — “Summer of ’69”, “007”.
- Calendar years are written as numerals — “1979”.
Spoken and non-word vocals#
- Spoken words — for example recorded dialogue from a documentary — must be transcribed.
- Sampled content must be transcribed, including non-word vocals, if it adds to the content of the song.
- Don’t include descriptions or names of screaming or sound effects.
Synced (time-stamped) lyrics#
You can sync your lyrics to the song by adding a timestamp before each line during release creation, in this exact format:
[00:00.000] Text of the line
└ [Minutes:Seconds.Milliseconds]
The format must be followed strictly:
- exactly one space between the closing bracket and the text;
- seconds and milliseconds separated by a dot (not a colon);
- milliseconds always three digits;
- no blank lines between lines of text, and none after the last line.
Common mistakes#
[00:09.007] Hey text ← two spaces after the bracket
[00.11.700] Hey text ← dot after minutes (must be a colon)
[00:15.013]Hey text ← no space after the bracket
[00:19.960] Hey text
← blank line after the last line
Correct version#
[00:09.007] Hey text
[00:11.700] Hey text
[00:15.013] Hey text
[00:19.960] Hey text
Whether Apple Music displays time-synced lyrics for a release is decided by Apple’s business and editorial priorities, so it can’t always be applied — but our support team can send a request. In that case, send the synchronised lyrics as a .ttml file to support. Apple documents the format in its Overview of TTML for Lyrics [CONFIRM: link to Apple’s TTML lyrics documentation]. If the request fits Apple’s internal priorities, the synced lyrics are applied.
For further detail, consult the lyrics section of the iTunes / Apple Music Style Guide. [CONFIRM: link to the Apple Music Style Guide lyrics section]
Thanks — your feedback helps us improve this article.
Still stuck?
Our support team answers tickets every business day. Include your account email and the release UPC or ISRC so we can jump straight in.