Cloud Spanner Unique index violation

I am using Cloud Spanner and am aware that one can interleave multiple tables eg: Singers > Albums > Songs (Albums is interleaved in Singers and Songs is interleaved in Albums). These tables have the primary keys singerId, albumId, songId respectively. I will keep this example to phrase my question better.

I create a new song called song1 which is from album1 by singer1.

Then when I try to create another song called song1 from album1 with singerId singer2, I get the error:

 Query failed: Unique index violation on index IDX_Songs_songId_U_1A1B6C1202613A13 at 
 index key [song1,singer1,album1]. It conflicts with row [song1,singer1,album1] in table
 Songs.

Does the new Song entry need to have unique keys within the Songs and Albums tables even though the outermost singerId key is different?

In short, I want to know why [singer1,album1,song1] and [singer2,album1,song1] are seen as duplicates even though their parent key differs.

The minimal DDL is given below:

Singers

CREATE TABLE Singers (
  
  singerId STRING(MAX) NOT NULL,

  firstName STRING(50) NOT NULL,

  surname STRING(50) NOT NULL,

  status INT64,

) PRIMARY KEY(singerId);

Albums

CREATE TABLE Albums (

  singerId STRING(MAX) NOT NULL,

  albumId STRING(MAX) NOT NULL,

  releaseDate DATE,

  genre BOOL,

) PRIMARY KEY(singerId, albumId),

  INTERLEAVE IN PARENT Singers ON DELETE CASCADE;

Songs

CREATE TABLE Songs (

  singerId STRING(MAX) NOT NULL,

  albumId STRING(MAX) NOT NULL,

  songId STRING(MAX) NOT NULL,

  duet BOOL,

) PRIMARY KEY(singerId, albumId, songId),

  INTERLEAVE IN PARENT Albums ON DELETE CASCADE;