LRC lyrics

OpenSavePlain textLyricsKaraoke

The lyrics file music players scroll in time with a song: each line after the moment it is sung. Enhanced LRC adds the moment of each word, for karaoke. Subtitld opens both, keeps the word timing, and writes it back.

At a glance

Extension.lrc
Media typeNone registered
SpecificationNone. Lyrics Displayer introduced it in 1998; the A2 Media Player added word times, as enhanced LRC.
Time precisionHundredths of a second; some files write thousandths
Text encodingNot declared. Subtitld detects it, and writes the encoding you choose.
StylingNone
PositioningNone
SpeakersNone. The Walaoke extension marks lines M:, F: or D: for duets.
In containersNone. A file beside the song, with the same name.

Anatomy of a file

Tags about the song, then a line of lyrics after each moment. What Subtitld writes for two subtitles, the first with word timing:

[ti:Tears of Steel]
[ar:The Blender Foundation]
[al:Open Movies]
[by:Celia]
[00:12.40]<00:12.40>Hold <00:12.80>on <00:13.10>to <00:13.30>the <00:13.90>light<00:14.90>
[00:14.90]
[00:15.20]We remember every word
[00:17.80]
  1. 1 · ID tags

    What the song is: its title, artist and album, and who made the file.

  2. 2 · Moment

    When the line is sung. A moment with nothing after it clears the line.

  3. 3 · Words

    The line, all on one. In enhanced LRC, each word after its own moment, the last moment where the last word ends.

The timecode

No hours: minutes go past 59, and past 99 in a long recording. Some files write thousandths, a colon before the fraction, or no fraction at all.

ID tags

Tag What it holds In Subtitld
[ti:] The song's title Kept
[ar:] The artist Kept
[al:] The album Kept
[au:] Who wrote the song Kept
[by:] Who made the LRC file Kept
[length:] How long the song is Kept
[offset:] Milliseconds to take from every moment; a positive one shows the lines sooner Applied as the file opens
[re:], [ve:] The program that wrote the file, and its version Not kept
[#:] A comment Not kept

Lines, and words

A line is shown until the next moment. A line sung more than once carries all its moments: [00:21.10][01:12.40]Chorus.

Enhanced LRC puts a moment before each word, in angle brackets, inside the line. A word lasts until the next stamp.

LRC Enhanced LRC
A moment Each line's, [01:12.40] Each line's, and each word's, <01:12.40>
Where a line ends At the next moment At the next moment
Where a word ends — At the next word stamp
Read by Most music players Players with karaoke, and YouTube

Word timing in Subtitld

Subtitld keeps the moment of each word in the subtitle's text, as ASS writes karaoke: a code before each word, its length in hundredths of a second.

{\k40}Hold {\k30}on {\k20}to {\k60}the {\k100}light

The codes show in the subtitle's text; the preview shows the words alone. Exported as ASS, the timing is ASS karaoke; exported as LRC, it is word stamps again.

A .usfx project does not keep word timing yet. Saved as a project, a subtitle keeps its first code and loses the rest. Export LRC or ASS to keep it.

LRC in Subtitld

Kept in the .lrc
  • Every subtitle's text, its lines joined on one
  • The moment of each word, as word stamps
  • Start, and an end where a pause follows, to the hundredth
  • The title, artist, album, songwriter, maker and length
Nowhere to go
  • Line breaks inside a subtitle
  • Italic, bold, underline and colour
  • Where a subtitle sits on the screen
  • Speakers' names as such, translations beside the original
  • Dubs, speakers' faces and colours

Save a .usfx project to keep them.

Opening an LRC file

Subtitld reads:

  • a subtitle a line and moment, a line with several moments once for each;
  • each line as lasting until the next moment; the last one until its last word stamp, else three seconds;
  • hundredths, thousandths, a colon before the fraction, or none;
  • [offset:], on lines and words alike;
  • word stamps as word timing, with or without spaces around them, as the A2 player writes them;
  • the lines in order of time, whatever order the file has them in;
  • any encoding.

Walaoke's M:, F: and D: stay in the text. A .lrc with no line that starts with a moment says so.

Exporting LRC

Subtitld's export dialog with LRC chosen, speaker names left out, formatting kept, UTF-8 and Linux line endings

Export, then Subtitles and LRC, offers:

  • TextThe original, a translation, or both, the translation after it on the same line.
  • Speaker namesLeft out, or before the text (Thom: …).
  • FormattingKept, or removed. Removed, the word timing goes with it, and each line is written plain.
  • Shift timesEvery subtitle earlier or later by the same amount.
  • EncodingUTF-8, with or without a byte order mark, UTF-16, or a Windows code page: Western European, Central European, Cyrillic or Korean.
  • Line endingsLinux and macOS, or Windows.

Subtitld writes the ID tags the project has, then a line for each subtitle at its start, in hundredths, with no offset. Where a subtitle ends before the next begins, or is the last, a moment with nothing after it says so. Empty subtitles are left out.

Trivia

1998

Kuo (Djohan) Shiang-shiang's Lyrics Displayer, from Taiwan, brought the format out, with a plug-in for Winamp.

Enhanced

YouTube lists LRC among the caption files it takes: no markup recognised, but enhanced LRC supported.

M: F: D:

Walasoft's Walaoke marks each line of a duet as the man's, the woman's or both, to show them in different colours.

Related formats

Have a .lrc to fix?

Open it in Subtitld with the video and retime it on the waveform.

Download Subtitld