Lilypond: How to avoid overshooting melisma in last stanzas [incl. MWE and output]

Question

I am trying to create some sheet music with lyrics. Some features of the music are:

• There are repeat sections in the music with alternative endings.
• The music starts on the chorus.
• There are more than one verse.
• The lyrics contain melisma (single syllable sung over several notes)

For some strange reason the melisma overshoots in the second and third verse, and I cannot seem to understand why this occurs. Is this a bug, or can this somehow (elegantly) be avoided? It almost seems like it overshoots to the first note in the third alternative ending...

I have spent some hours scavenging the manual, but can't seem to find an explanation for this. Any help is greatly appreciated! Please also point out any bad practice, as I am new with this software.

Please see output and corresponding MWE below.

\version "2.24.0"

global = {
  \key f \major
  \time 4/4
}

sopranoVoice = \relative c'' { \new Voice = "stemme" {
  \global

  % Music follows here.
  \repeat volta 3 {
\bar ".|:" c4 4 4 4 |
 \alternative{
 \volta 1,2 
 {d4 4 4 4|
 g,4 4 4 4 | 
 c,4 e g bes16~ 16~ 16~ 16|}
\volta 3
{ e4 4 4 8 8|}
 }%end alternatives
  }% end repeats
}% endVoice
} %end music


verseOneSopranoVoice = \lyricmode {
  \set stanza = "Chorus:"
  % Lyrics follow here.
  Chorus! is my desire, also part of chorus!
  <<{\set stanza = "1." 
  Now I sing my be -- au -- ti -- ful __
    }
    \new Lyrics 
    {\set associatedVoice = "stemme"
      \set stanza = "2." Oh how wonder -- ful is this happy song! __
    }
    \new Lyrics 
    {\set associatedVoice = "stemme"
      \set stanza = "3." Oh yes, there! I see over there is __
    }
    >>
  and the song is over.
}


voiceParts = \new ChoirStaff << 
\new Staff  { \sopranoVoice }
\addlyrics { \verseOneSopranoVoice }
>>

\score {
  <<
    \voiceParts
  >>
  \layout { }
}

Answer 1

Reposting my message to the mailing list here for transparency:

This behaviour is caused by defining new Lyrics contexts in Lyrics contexts as it seems. I’ve also never seen someone use such notation before.

Rather I’d do different Voices for different Lyrics to align to:

\version "2.24.0"

global = {
\key f \major
\time 4/4
}

sopranoVoice = \relative c'' {
  \new Voice = "chorus" {
  \global
  
  % Music follows here.
  \repeat volta 3 {
    \bar ".|:" c4 4 4 4 |
    \alternative{
      \volta 1,2
      { d4 4 4 4|
        \new Voice = "stanza" {
          g,4 4 4 4 |
          c,4 e g bes16~ 16~ 16~ 16|
        }
      }
      \volta 3
      { e4 4 4 8 8|}
      }%end alternatives
    }% end repeats
  }% endVoice
} %end music


verseOneSopranoVoice = <<
  \new Lyrics \lyricsto "chorus" {
    \set stanza = "Chorus:"
    % Lyrics follow here.
    Chorus! is my desire, also part of  \set associatedVoice = "stanza" 
chorus!

    \set stanza = "1."
    Now I sing my be -- au -- ti -- \set associatedVoice = "chorus" ful __
    
    and the song is over.
  }
  \new Lyrics \lyricsto "stanza"
  {
    \set stanza = "2." Oh how wonder -- ful is this happy song! __
  }
  \new Lyrics \lyricsto "stanza"
  {
    \set stanza = "3." Oh yes, there! I see over there is __
  }
>>

voiceParts = \new ChoirStaff <<
  \new Staff { \sopranoVoice }
  \verseOneSopranoVoice
>>

\score {
  <<
    \voiceParts
  >>
  \layout { }
}