Dhageyso ka qaybgal oo higaad Kediska

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Waa maxay labada qaybood ee ugu muhiimsan qaabka LAS?
   • Dhageyste iyo Higaad
   • Encoder iyo tokenizer
   • Dab-dhaliye iyo takoorid
   • Shaandheeye iyo kala saar
2. Muxuu sameeyaa Higaada ku jirta LAS?
   • Taleefannada qaamuuska
   • Jilayaasha, midba mar
   • Erayada oo dhan kaliya
   • Muuqaalka soo noqnoqda
3. Waa maxay sababta koodka LAS loogu yeedhay 'pyramidal'?
   • Waxay shaashadda u qaabaysan tahay saddex xagal
   • Waxay isticmaashaa saddex GPUs
   • Waxay yaraynaysaa xallinta wakhtiga lakab kasta
   • Waxay haramaysaa qiyamka khasaaraha
4. Waa maxay qaybta da'da weyn ee LAS gaar ahaan AANU u baahnayn?
   • makarafoon
   • Shaqada khasaaraha
   • Xogta tababarka
   • Qaamuuska dhawaaqa telefoonka ee gacanta lagu dhisay
5. Farsamaynkee u ogolaanaya Higaada inuu diirada saaro qaybta ku habboon ee maqalka ee jile kasta?
   • Fiiro gaar ah
   • Ka saarida
   • Dufcaddii caadiga ahayd
   • Manjooyin beam ah
6. Waa maxay hal xaddidaadda qaabka asalka ah ee LAS?
   • Ku shaqayn kari wayday qalab kasta
   • Waxa ay ahayd mid aan soconayn, oo u baahan odhaah buuxda ka hor inta aan koodka la saarin
   • Kaliya waxay soo saartaa tirooyinka
   • Waxay u baahday qaamuus telefoonme ah
7. What WER did the LAS paper report on its Google voice-search subset without a language model?
   • 1.41%
   • 41.0%
   • 14.1%
   • 80.0%
8. What distinction does the LAS paper make about dependencies among output characters?
   • It predicts each character independently with no sequence model
   • It outputs only phoneme classes and forbids character context
   • It requires a separate dictionary to define character dependencies
   • The model emits character sequences without assuming the characters are independent