Get a song’s analysis
get_song_analysisWord-timed lyrics, tempo, beats, sections, hook and the parts a video can be cut to parts lists where a video can start and end without cutting a word: each section (Verse 1, Chorus 1…), clips of about 15 seconds cut between the lines of long sections (kind "clip", parent: its section) and each verse with the chorus after it (kind "pair"); pass a part’s start and end as a video’s trim (start_seconds, end_seconds), picking one at least as long as the video type’s minimum (a lyric video 8 seconds, a music video 15). default_part is the one the app starts on: the sung part nearest 15 seconds. lyrics.text is the song’s lyrics to edit: yours as you gave them, or the transcription a line per phrase (lyrics.yours says which). The analysis also listens for who sings: map.voices lists where someone other than the lead artist sings (voice "other": another solo voice; "group": a choir, a crowd or backing singers), and those words carry backing: true (music videos keep them off camera, never lip-synced by the artist); map.instrumental is true when no vocals were heard. A song analyzed before voices were heard has no map.voices until its first music video plan listens to it.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"song_id": "<song_id>"
},
"name": "get_song_analysis"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| song_id | Yes | The song id | |
| min_seconds | No | The parts for a video at least this long: a music video 15 (its clips are cut no shorter). Default 8, a lyric video |