Idanwo ti o sopọ mọ itọsọna · Lile Ipele

Ti o ku Vector Quantization adanwo

Test residual vector quantization, including iterative error coding, compact codebooks, commitment loss, and bitrate-quality control.

Jẹmọ itọsọna awọn ọnaTi o ku Vector Quantization
Ibeere 1 ti 8

Kini iwe koodu ti o tẹle kọọkan ni iwọn RVQ?

Awotẹlẹ gbogbo ibeere8 ibeere

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

  1. Kini iwe koodu ti o tẹle kọọkan ni iwọn RVQ?
    • Aṣiṣe ti o ku silẹ nipasẹ awọn ipele ti tẹlẹ
    • Awọn atilẹba input lẹẹkansi lati ibere
    • Ikanni ohun afetigbọ ti o yatọ
    • Awọn ti o wu decoder
  2. Kilode ti o lo RVQ dipo iwe koodu nla kan ṣoṣo?
    • O yago fun iwulo fun ikẹkọ eyikeyi
    • O ṣe agbekalẹ iwe koodu ti o munadoko nla sinu awọn kekere, fifipamọ iranti ati iṣiro
    • O ṣe imukuro kooduopo patapata
    • O ṣiṣẹ nikan fun data ọrọ
  3. Bawo ni atunkọ ipari ti a ṣe lati awọn koodu RVQ?
    • Nipa gbigbe nikan kẹhin codebook ká fekito
    • Nipa aropin igbewọle pẹlu ariwo
    • Nipa akopọ awọn fekito ti o yan lati gbogbo awọn ipele
    • Nipa isodipupo gbogbo awọn titẹ sii codebook
  4. Pẹlu awọn ipele M ti awọn titẹ sii K kọọkan, melo ni awọn akojọpọ koodu ti o munadoko ti RVQ ṣe aṣoju?
    • M igba K
    • M pẹlu K
    • K pin nipasẹ M
    • K si agbara M
  5. Kini idi aṣoju ti 'pipadanu ifaramo' nigbati ikẹkọ RVQ codebooks?
    • Lati tọju awọn abajade koodu koodu sunmọ awọn titẹ sii iwe koodu koodu ti wọn yan
    • Lati mu iwọn ayẹwo ohun pọ si
    • Lati pa awọn fireemu iwe ohun ti ko lo
    • Lati yi awọn koodu pada si ọrọ
  6. Bawo ni RVQ ṣe jẹ ki kodẹki nkankikan kan yi bitrate rẹ pada ni akoko itọkasi?
    • Nipa yiyipada si oṣuwọn ayẹwo ti o yatọ nikan
    • Nipa lilo diẹ ẹ sii tabi diẹ si awọn ipele quantizer
    • Nipa atunṣe gbogbo awoṣe ni akoko kọọkan
    • Nipa compressing awọn iwọn decoder
  7. In RVQ training, what does commitment loss encourage?
    • Each codebook to contain a different number of entries
    • The decoder to output longer audio
    • The encoder output to stay close to the selected codebook vector
    • The residual to increase at every quantization stage
  8. How are codebook vectors commonly updated in the guide’s RVQ description?
    • By sorting the codebook alphabetically
    • By replacing the codebook with a recurrent network
    • By adding the residual directly to every code index
    • With exponential moving averages of encoder assignments