Sung line

Sung "hold on to me", confirmed in the audio, with long held vowels and visible breath.

Sung "hold on to me", confirmed in the audio, with sustained open vowels and visible breath.

Sung "hold on to me", confirmed in the audio, with sustained open vowels and visible breath.

Sings the requested lyric with mouth shapes convincingly synced to the sustained vocal notes.

Audio present, sustained melody matches the sung phonemes.

Sustained vowel shapes and visible breath match the sung phrase precisely.

Audible singing with sustained vowel shapes.

Audible singing with sustained, elongated vowel shapes distinct from normal speech.

Sung line matches, articulation tracks the melody.

Sustained vowels matched to mouth shapes, visible breath.

Sings "hold on to me" twice, confirmed in the audio, with sustained vowels; the clip cuts to a close-up of the mouth and back. Checked frame by frame.

Silent clip (no native audio); expressive mouth/vowel shapes plausible for studio singing.

Mouth shapes read as breathing/exhaling into cold air rather than singing vowel shapes -- the model appears to have depicted the wrong action entirely; not independently frame-verified, moderate-severe discount given how specific and confident the description was.

Mouth shapes and breath vapor present, but silent, no vocal track.

FAILURE MODE: no audio track; mouth shows generic singing-style morphs unrelated to the actual sung phrase.