Gis ndoorte ci quiz audio

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

1. Lan mooy 'onset' ci audio?
   • Saa yu xew-xew akustik tàmbalee, melni sonngug note
   • Fade-out bu ab son
   • Kawewaay bu bari bi ci benn track
   • Jeexal benn way
2. What does spectral flux measure for onset detection?
   • A rule that marks every beat as a note onset
   • Positive increases in spectral energy between successive frames
   • A decay detector that selects only falling energy
   • A pitch estimator that finds the nearest MIDI note
3. Lan moo waral ñuy jëfandikoo genn-wàllu vague ci flux spectral?
   • ngir yokk decay yi
   • Ngir soppi audio ci mbind
   • Kon coppite energie yuy yokk (positif) kese ñoo ciy bokk, bàyyi xel ci yàqu-yàqu yi
   • ngir yeexal audio bi
4. Ban xeetu ndoorte moo gëna am jafe-jafe gis?
   • Tambour bu ñaw
   • Cymbal bu am doole
   • backbeat bu fiir
   • Violon buy woyof ndànk
5. Lan la etape picking bi am threshold buy méngoo ak limuy tere?
   • Ñaari yoon ñuy tàmbali ak njuumte
   • Kompresioŋ audio
   • Coppite ci ton
   • Yaatu stéréo
6. What can a neural onset detector learn from labeled audio examples?
   • A guarantee that every onset will be detected
   • Patterns in the audio features that signal event beginnings
   • A fixed tempo that applies to every recording
   • The identity of every instrument without evaluation
7. What does onset backtracking do when detected times are used to create audio slices?
   • Moves each onset to the end of the entire recording
   • Replaces each onset with the nearest musical beat
   • Moves an onset to the nearest preceding local minimum of an energy signal
   • Adds a fixed one-second delay after every peak
8. What can happen when an onset model is transferred to a different instrument or dataset?
   • Its predictions are guaranteed to remain identical
   • The audio automatically becomes monophonic
   • The model no longer needs a time-frequency input
   • Its performance may change and should be evaluated in the new domain