Guide-yakabatana mibvunzo · Pakati Nhanho

Conformer Architecture Quiz

Learn how the Conformer combines convolution and self-attention for speech, including its block design and reported benchmark results.

Nzira dzekutungamira dzinoenderanaConformer Architecture
Mubvunzo 1 ye 8

Ndeapi maitiro maviri anosanganisirwa neConformer architecture mubhuroko rimwe chete?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Ndeapi maitiro maviri anosanganisirwa neConformer architecture mubhuroko rimwe chete?
    • Convolution uye kuzvidzora
    • Kudzokorora uye kubatana
    • Sarudzo miti uye kubatanidza
    • Beam yekutsvaga uye tokenization
  2. Sei kubatanidza convolution uye kutarisa kuchinyanya kubatsira pakutaura?
    • Tarisiro inobata ruzivo rwemafoni emunharaunda ukuwo convolution inobata mamiriro epasi rose
    • Convolution inobata maitiro emunharaunda ukuwo kuzvitarisa kunotora mamiriro enguva refu
    • Zvese zviri zviviri zvine basa remifananidzo, kwete odhiyo
    • Hazvibatsiri; ndezvekumhanya chete
  3. Chii chinonzi 'Macaron' kana sandwich chimiro cheConformer block?
    • Matanho maviri ekutarisisa akatenderedza imwe convolution
    • Imwe feed-forward layer chete
    • Kutarisa uye convolution yakaputirwa pakati pemahafu-nhanho feed-forward modules
    • Convolution yakaturikidzana pamusoro pe convolution
  4. Isangano ripi rakasuma Conformer, uye mugore ripi?
    • OpenAI muna 2018
    • Meta muna 2015
    • Microsoft muna 2012
    • Google muna 2020
  5. Ko iyo convolution module mukati meConformer inowanzo sanganisira chii?
    • A pointwise conv, iyo GLU, yakadzama conv, normalization, uye Swish activation
    • Ingori imwe chete yakanyatsobatanidzwa layer
    • Sero rinodzokororwa reLSTM
    • A random forest classifier
  6. On LibriSpeech test-clean/test-other, what WER did the original Conformer paper report with an external language model?
    • 2.1% / 4.3%
    • 1.9% / 3.9%
    • 9.1% / 9.3%
    • 0.1% / 0.3%
  7. What positional information does the Conformer paper add to attention?
    • A geographic GPS coordinate for every speech frame
    • A fixed class label for each utterance
    • Relative positional encoding
    • A separate video frame index
  8. What word error rates did the Conformer paper report without an external language model on LibriSpeech test/test-other?
    • 20.1% / 40.3%
    • 0.1% / 0.3%
    • 12.1% / 14.3%
    • 2.1% / 4.3%