Singing Voice Synthesis Quiz

Answers and explanations appear as you take the quiz.

1. What core inputs does a score-conditioned singing-voice system use?
   • Lyrics and musical-score information such as pitch and duration
   • Only a text prompt describing the mood
   • A photo of the singer and a genre label
   • A drum pattern and a chord chart
2. What did Yamaha say it announced publicly in 2003?
   • A diffusion-based speech recognizer
   • Development of a singing-synthesis technology called VOCALOID
   • The first open-source music transformer
   • A universal audio codec for streaming
3. What does an F0 contour represent in voiced singing?
   • The loudness envelope of the song
   • The lyrics' phoneme order
   • The estimated fundamental-frequency or pitch trajectory over time
   • The reverb level of the room
4. In the described two-stage neural singing-synthesis setup, what can the acoustic model produce before the vocoder?
   • A finished MP3 file
   • A MIDI file
   • A text transcript
   • A mel-spectrogram
5. How does DiffSinger use diffusion in its acoustic model?
   • It iteratively denoises acoustic features conditioned on the musical score
   • It concatenates complete songs without conditioning
   • It predicts only the singer’s name and no audio features
   • It replaces note information with an image classifier
6. Which is a major ethical concern raised by modern SVS?
   • It uses too much sheet music
   • Deepfake vocals of real artists made without consent
   • It cannot produce any pitch variation
   • It only works for spoken word
7. What does portamento describe in a synthesized singing performance?
   • The word-level transcript of lyrics
   • The codec that converts a waveform into tokens
   • A continuous pitch glide between notes
   • A fixed speaker identity embedding
8. What does a neural vocoder do after an acoustic model predicts a mel-spectrogram?
   • Chooses the lyrics and musical score
   • Determines the singer’s legal permissions
   • Estimates the track’s album cover
   • Converts the acoustic representation into an audio waveform