Guide-yakabatana mibvunzo · Pakati Nhanho

MelGAN Generative Vocoder Quiz

Nzwisisa kuti MelGAN inomepu sei mel-spectrograms kune waveforms, inoshandisa convolutional upsampling, uye inodzidzisa vanosarura pazvikero zvakawanda.

Nzira dzekutungamira dzinoenderanaMelgan Generative Vocoder
Mubvunzo 1 ye 8

Ndeipi yekuisa iyo MelGAN inoshandura kuita yakaomeswa waveform?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Ndeipi yekuisa iyo MelGAN inoshandura kuita yakaomeswa waveform?
    • A mel-spectrogram
    • Mavara emavara manyoro
    • A MIDI faira
    • Rutivi rwephoneme
  2. Nei MelGAN ichikurumidza kupfuura WaveNet pakufungidzira?
    • Inoshandisa chiyero chidiki chemuenzaniso
    • Iyo isiri-autoregressive uye inogadzira odhiyo mune imwe yakafanana pass
    • Inomhanya chete paTPUs
    • Inodarika mabhendi epamusoro-frequency
  3. Ndeipi dhizaini yerusarura yakaunzwa naMelGAN?
    • Mumwe musarura wepasi rose
    • A transformer-based discriminator
    • A multi-scale discriminator anoshanda pane odhiyo pane akati wandei sarudzo
    • Hapana kusarura zvachose
  4. Ndekupi kurasikirwa kunobatsira kudzikamisa kudzidziswa kwaMelGAN?
    • Muchinjikwa-entropy pane zvinyorwa zvinyorwa
    • Kurevesa pfungwa zvibodzwa kurasikirwa
    • Triplet margin kurasikirwa
    • Feature-kufananidza kurasikirwa pakati pekusarura activation
  5. Jenareta yeMelGAN inowedzera sei munda wayo wekugamuchira?
    • Nekushandisa dilated convolutions mumabhuroko akasara
    • Nekuwedzera akadzokororwa LSTM maseru
    • Nekushandisa kutarisisa pane clip yese
    • By stacking pooling layers
  6. What subjective evaluation measure did the MelGAN paper report?
    • Shoko kukanganisa mwero (WER)
    • Mean Opinion Score (MOS)
    • BLEU score
    • Peak signal-to-noise ratio (PSNR) only
  7. What do MelGAN’s transposed convolutions do in the generator?
    • Convert the waveform back into text
    • Choose the phonemes in the input sentence
    • Upsample the mel-spectrogram representation toward the audio waveform rate
    • Reduce the number of audio samples to zero
  8. Which audio domains did the MelGAN paper use to illustrate the generality of its techniques?
    • Machine translation, image captions, and video subtitling
    • Only speech recognition and speaker identification
    • Text summarization, code generation, and image classification
    • Speech synthesis, music-domain translation, and unconditional music synthesis