Kuiz berkaitan panduan · Keras Tahap

Kuiz Bisikan Penalaan Halus

Rancang penyesuaian Whisper menggunakan pasangan teks audio yang dijajarkan, pemisahan pembesar suara yang bersih, pertukaran saiz model dan penilaian yang mantap untuk melupakan.

Laluan panduan yang berkaitanBisikan Penalaan Halus
soalan 1 daripada 8

What changes during supervised fine-tuning of Whisper?

Pratonton setiap soalan8 soalan

Jawapan dan penjelasan muncul semasa anda mengambil kuiz.

  1. What changes during supervised fine-tuning of Whisper?
    • Only the decoded transcript is edited after inference
    • Only the language label is changed without training
    • Model weights are updated using paired audio and text examples
    • Only a static list of spellings is appended to the output
  2. Why must audio and transcript segments be aligned?
    • Alignment guarantees the same GPU memory use for every segment
    • Misaligned text teaches the model incorrect speech-to-text mappings
    • An alignment error is harmless if the transcript contains domain vocabulary
    • Correct timing guarantees that every transcript word is accurate
  3. To assess performance on unseen speakers when each speaker contributes many recordings, how should the data be partitioned?
    • Randomly split overlapping windows from each speaker
    • Place every speaker in every partition
    • Use test speakers during training only
    • Group by speaker before creating train, validation and test splits
  4. What does a held-out validation set support during fine-tuning?
    • Checkpoint and hyperparameter decisions
    • Unlimited tuning followed by an unbiased final score on that same set
    • Gradient updates using every held-out label
    • A guarantee that the final deployment distribution will be unchanged
  5. What can differ between two word error rate reports?
    • Only the number of model parameters
    • Only the loss observed on the training set
    • Text normalization and tokenization conventions
    • Only whether both systems use the same tokenizer name
  6. Why monitor a general-domain set while adapting to a narrow domain?
    • General data always lowers target accuracy
    • Target adaptation can reduce performance outside its training distribution
    • The test set should be used for gradient updates
    • The model size changes automatically
  7. What can low-rank adapters offer when the selected implementation supports them?
    • Guaranteed lower WER in every domain
    • Automatic correction of mislabeled transcripts
    • A guarantee that no general-domain behavior can change
    • Fewer trainable parameters during adaptation
  8. Why compare fine-tuning with decoding adjustments or domain-term support?
    • Training may add cost when simpler changes solve the problem
    • Decoding choices always change model weights
    • Fine-tuning cannot affect vocabulary behavior
    • A baseline makes test labels unnecessary