Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/6xingyv/accompanist-lyrics-core/llms.txt

Use this file to discover all available pages before exploring further.

Standard LRC is the most widely supported lyrics format. Each line of lyrics is prefixed with a [mm:ss.xx] timestamp, and optional ID3-style header tags carry metadata such as the song title, artist, and album. EnhancedLrcParser handles both standard LRC and the Enhanced LRC extension from a single object—you do not need to choose between them.

​
Format

A well-formed LRC file looks like this: 
[ti:Song Title]
[ar:Artist Name]
[al:Album Name]
[offset:500]
[00:12.34]First line of lyrics
[00:12.34]Translation of first line
[00:23.45]Second line of lyrics
[00:35.00]Third line

​
Metadata tags

The following header tags are recognised and surfaced on the returned SyncedLyrics object:
TagFieldNotes
[ti:…]SyncedLyrics.titleSong title
[ar:…]SyncedLyrics.artistsSlash-separated artist names; colon separates role from name
[al:…](parsed but not stored on the model)Album name
[offset:…](parsed but not applied)Signed integer millisecond offset; recognised but not used to shift timestamps
[length:…](parsed but not stored)Total track duration as an integer in milliseconds

​
Multiple artists

The [ar:] tag can hold several artists separated by /. An optional role prefix separated by : is also supported: 
[ar:Lead Singer/Featuring Artist]
[ar:Main:Lead Singer/Feature:Featuring Artist]
Each entry becomes an Artist(role, name) in SyncedLyrics.artists.

​
Translation lines

A translation is encoded as a second line that shares the same timestamp as its primary line but carries different text content. The parser detects this automatically and stores it in SyncedLine.translation: 
[00:12.34]Hello world
[00:12.34]你好世界
Both lines parse to a single SyncedLine where content = "Hello world" and translation = "你好世界".

​
Detection

EnhancedLrcParser considers content parseable when it contains at least one [dd:dd.dd] (or [dd:dd.ddd]) timestamp pattern: 
content.contains("""\[\d{2}:\d{2}\.\d{2,3}\]""".toRegex())

​
Usage

EnhancedLrcParser is a Kotlin object (singleton)—no instantiation required: 
// Parse from a raw string
val lyrics = EnhancedLrcParser.parse(lrcContent)

// Parse from a pre-split list of lines
val lyrics = EnhancedLrcParser.parse(lrcLines)

​
Accessing the result

val lyrics = EnhancedLrcParser.parse(lrcContent)

println(lyrics.title)                        // "Song Title"
println(lyrics.artists.firstOrNull()?.name)  // "Artist Name"

for (line in lyrics.lines) {
    if (line is SyncedLine) {
        println("${line.start}ms: ${line.content}")
        line.translation?.let { println("  → $it") }
    }
}
SyncedLine carries the following fields:
FieldTypeDescription
contentStringThe lyric text
translationString?Translation if a same-timestamp line was found
startIntStart time in milliseconds
endIntEnd time in milliseconds (set to the next line’s start)
When the LRC content also contains syllable timing brackets (<mm:ss.xx> or [mm:ss.xx] inside a timestamped line), EnhancedLrcParser produces KaraokeLine instances instead of SyncedLine. See the Enhanced LRC page for details.

​
Example: full parse

val lrcContent = """
    [ti:Starlight]
    [ar:Example Artist]
    [al:Demo Album]
    [00:10.00]Under the open sky
    [00:10.00]在开阔的天空下
    [00:15.50]I see the stars align
    [00:21.00]And fade into the night
""".trimIndent()

val lyrics = EnhancedLrcParser.parse(lrcContent)

println(lyrics.title)   // Starlight
println(lyrics.lines.size) // 3

val first = lyrics.lines[0] as SyncedLine
println(first.content)      // Under the open sky
println(first.translation)  // 在开阔的天空下
println(first.start)        // 10000

Build docs developers (and LLMs) love

Get started for freeTalk to us