quiz bu lëkkale ak tegtal · Dëgër Niveau

Quiz di tàggat yaram budul soppiku

Saytu naka la taggat permutation-invariant di doxalee ci génnug waxkat bu jaarul yoon, ak lu tax wax-niveau devoir am solo ci diiru tàqale waxkat yu bari.

Yooni njiit yu jëm ciPermutasioŋ tàggat yaram budul soppiku
Laaj 1 de 8

Ban jafe-jafe bu mag la tàggat yaram budul soppiku (PIT) di saafara?

Xoolaatal laaj bu nekk8 laaj

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

  1. Ban jafe-jafe bu mag la tàggat yaram budul soppiku (PIT) di saafara?
    • Xool ban chaine de sortie moo wara méngoo ak ban waxkat ci diiru tàggat bi
    • Komprime audio ngir denc
    • Wàññi njëgu jumtukaayi mikro
    • Tekki kàddu yuñ tàqale ci mbind
  2. naka PIT di xamee ban njuumte lañu wara jëfandikoo suñuy yeesal model bi?
    • Dafay xayma njuumte ci bépp génne bu tolloo
    • Dafay jéem bépp lëkkaloo diggante génn ak balluwaay ba noppi tëye benn bi am njuumte bi gëna ndaw
    • Saa yu nekk dafay jox baat bi gëna kawe ci génnug 1
    • Dafay baña bàyyi xel ci royuwaay yi
  3. Su benn pexe bu melni PIT amul, lan mooy xew ci génnug modelu tàqale bi?
    • Ñu daal di set ci saasi
    • Dañu gënoon daw
    • Defar nañu ay génne yu fësul, yu yam ndax etiketu liggéey yu wuute
    • Soxla wuñu benn done tàggat
  4. Ngir tàqale N waxkat, ñaata permutation la PIT wara bàyyi xel ci jéego bu nekk?
    • N (ligneer ci kàddukat yi)
    • N kaare
    • Saa yu nekk ñaar
    • N faktoriyaal
  5. Ban njariñ la PIT ci wàllu wax (uPIT) yokk ci wàllu liggéey ci wàllu kaadar?
    • Dafay tëye waxkat bi ci chaine de sortie bu dëgër ci wax ji yépp
    • Dina dindi bépp royuwaay
    • Music rek lay dox, du wax
    • Dafay yokk limu waxkat yi ñaari yoon
  6. Ban gis-gis moo mëna dindi jafe-jafe permutation bi?
    • Yokk bitrate audio bi
    • Extraksioŋ biy waxkat biñ bëgga wax, te dafay aju ci baat buy royuwaay
    • Jëfandikoo enregistrement mono
    • Desactiver répropagation
  7. What extra inference step does utterance-level PIT avoid compared with frame-level PIT?
    • Running speech recognition before source separation
    • Estimating the number of microphones in the recording
    • Solving an additional output-speaker permutation problem during inference
    • Recomputing the training loss after every generated word
  8. Which benchmark does the guide name as an example for evaluating speech-separation models?
    • LibriSpeech for single-speaker transcription
    • ImageNet for visual recognition
    • GLUE for language understanding
    • WSJ0-2mix for overlapping-speaker separation