Wav2Vec 2.0 Quiz

Answers and explanations appear as you take the quiz.

1. What core problem in speech recognition was Wav2Vec 2.0 designed to address?
   • The scarcity of labeled (transcribed) audio
   • Audio files being too large to store
   • Microphones being too expensive
   • Text being hard to tokenize
2. What kind of learning objective does Wav2Vec 2.0 use during pretraining?
   • Direct phoneme classification with full labels
   • A contrastive loss over masked latent segments
   • Pixel reconstruction
   • Reinforcement learning from human feedback
3. What component first processes the raw waveform in Wav2Vec 2.0?
   • A recurrent LSTM
   • A diffusion sampler
   • A multi-layer CNN feature encoder
   • A handcrafted MFCC extractor
4. In a reported LibriSpeech experiment, what small labeled-data setup demonstrated low-label potential?
   • 10 minutes unlabeled with no pretraining
   • 53,000 hours labeled with no pretraining
   • One hour labeled with no unlabeled data
   • 10 minutes labeled after pretraining on 53,000 hours of unlabeled speech
5. What is the role of the learned codebook in Wav2Vec 2.0?
   • It discretizes continuous audio into finite speech units
   • It stores user passwords
   • It compresses the final transcript
   • It selects the microphone gain
6. Which large multilingual model is built on the Wav2Vec 2.0 approach?
   • GPT-2
   • XLS-R
   • Stable Diffusion
   • ResNet-50
7. What does the Transformer contribute during Wav2Vec 2.0 pretraining?
   • Converts transcripts into audio
   • Clusters final text into language labels
   • Contextualizes latent speech features around masked spans
   • Removes the feature encoder
8. What is typically done after self-supervised pretraining to create an ASR model?
   • Replace the audio encoder with an image classifier
   • Train only on unlabeled text
   • Discard the model and use k-means alone
   • Fine-tune with transcribed speech