quiz bu lëkkale ak tegtal · Diggante Niveau

Quiz ci modelu tekkikatu RNN

Xam naka la xeetu RNN-Transducer di boolee kadre akustik ak jaar-jaaru génne ngir liggéeyi transkripsioŋ yu toppalante.

Yooni njiit yu jëm ciModèlu traducteur Rnn
Laaj 1 de 8

Ban yamaleg CTC la RNN-Transducer di wax ci?

Xoolaatal laaj bu nekk8 laaj

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

  1. Ban yamaleg CTC la RNN-Transducer di wax ci?
    • Tegtal moomel sa bopp bu CTC, biy tere modele dependency token yiy génne
    • CTC mënul def audio
    • CTC dafay laaj GPU ngir mëna dox
    • CTC du dox ludul ci nataal yi
  2. Lan mooy ñetti mbir yi gëna am solo ci RNN-Transducer?
    • Enkodeer, dekodeer, ak tànnal
    • Encoder (reso transkripsioŋ), reso buy wax luy am, ak reso buñ boole
    • Tokenisatër, dugalkat, ak softmax kese
    • Ñaari boppu ak benn routeur
  3. Ban cër la reso biy wax luy am ci RNN-T?
    • Dafay kompresse forme onde audio
    • Dafay natt volume mikro bi
    • Dafay melni xeetu làkk, di aju ci ay token yuñ njëkka génne
    • Dafay tànn tolluwaayu échantillonnage bi
  4. Lan moo waral RNN-T di jàppale streaming bu woyof ci anam wu natureel?
    • Dafay jëfandikoo kàddu yépp ci saasi balaa muy génne dara
    • Dafay baña bayyi xel ci doxalinu waxtu
    • Dafay dox ludul ci fichier yuñ njëkka enregistre
    • Liggéeyam dafa monotone ak cammoy-ndeyjoor, awaase ci audio ak ay bërëb yu amul dara
  5. Ci dekodaasu RNN-T, lan mooy génne token bu amul dara?
    • Dellu ci jéego bi ci topp ci waxtu audio
    • Jeexal transkripsioŋ bi yépp
    • efaase baat bi njëkk
    • Wecci ci beneen làkk
  6. Ban xeetu encoder mooy gëna wecci LSTMs ci sistem RNN-T yu bees yi?
    • Garab yiy jël dogal
    • Enkodeer konformer
    • Sac-de-mots
    • Modèle Markov yu nëbbu
  7. What two histories condition the RNN-T joint network’s output distribution?
    • Only the future transcript and a speaker photo
    • Only the last acoustic frame, without output context
    • The encoder representation of acoustic input and the prediction-network representation of prior output labels
    • The optimizer state and learning rate
  8. What alignment behavior does an RNN-T allow before advancing to the next audio frame?
    • It must emit exactly one label for every frame
    • It advances time only after a word-final token
    • It generates all output before reading any audio
    • It can emit zero or more nonblank labels at the current frame, then a blank advances time