Ikibazo gihuza ibibazo · Hagati Urwego

Amajwi Amajwi hamwe no Guhagararira Kwiga Ikibazo

Ongera usuzume uburyo gushyiramo amajwi bifata amajwi, uburyo amahugurwa atandukanye ahuza amajwi ninyandiko, nuburyo ibice byize byimurwa mubikorwa.

Inzira zijyanye no kuyoboraKwinjiza amajwi no Kwiga Guhagararira
Ikibazo 1 Bya 8

Gushyira amajwi ni iki?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Gushyira amajwi ni iki?
    • Uburebure-burebure bwimibare yerekana amajwi yerekana ibintu bifatika
    • MP3 ya compression ya verisiyo yamajwi
    • Inyandiko yanditse yamagambo yavuzwe
    • Urwego rwijwi rwapimwe muri décibel
  2. Ni ukubera iki kwigira wenyine kwigira ari ngombwa cyane mugushira amajwi?
    • Birasaba amajwi kwandikwa muri studio
    • Yiga ibyerekanwa byingirakamaro bivuye kumajwi adashyizweho ikimenyetso, bigabanya gukenera ibirango byabantu
    • Ikora kuri muzika gusa, ntabwo ari imvugo
    • Iremeza neza neza inyandiko-mvugo
  3. Nigute Wav2Vec 2.0 yiga mugihe cyo kwitegura?
    • Muguhindura amajwi mu gifaransa
    • Mufata mu mutwe buri dosiye y'amahugurwa neza
    • Muguhisha amajwi yama majwi no guhanura ibice byuzuye mubirangaza
    • Mugupima ingano ya dosiye ya buri clip
  4. Niki icyitegererezo nka CLAP gihuza umwanya uhuriweho?
    • Amakadiri abiri atandukanye
    • Pixel hamwe namabara yamateka
    • Urwego rwa Bateri hamwe nikoreshwa rya CPU
    • Ijwi kandi rihuye nibisobanuro byanditse
  5. Mbere yo kugaburira amajwi urusobe rw'imitsi, ni izihe mpinduka zisanzwe zikoreshwa?
    • Guhindura kuri spekrogramu cyangwa yize muyunguruzi
    • Gucapura nkumuziki wimpapuro
    • Gusiba imirongo yose iri hejuru yo kumva kwabantu burundu
    • Guhindura ishusho ya JPEG yerekana igifuniko cya alubumu
  6. Ni ikihe gikorwa nyacyo-isi gishingiye ku kugereranya amajwi?
    • Kubara igiciro cyindirimbo
    • Speaker diarisation, kumenya uwabavuze igihe
    • Guhitamo alubumu yatwikiriye ibihangano
    • Gushiraho ecran ya ecran
  7. During contrastive audio-text training, what happens to a matching audio clip and its text description?
    • They are converted into the same raw waveform
    • The text embedding is discarded
    • Their embeddings are pulled closer while mismatched pairs are pushed apart
    • The audio is assigned a speaker name as a class label
  8. Why can one self-supervised audio embedding be reused across several downstream tasks?
    • It stores the original waveform without compression
    • It contains a manually written label for every possible sound
    • It works only for one speaker and one recording
    • It captures general acoustic structure that can support tasks with little extra labeled data