NCA-GENM domain - 15% of the exam

Multimodal Data

Multimodal Data is 15% of the NVIDIA-Certified Associate: Generative AI Multimodal (NCA-GENM) exam. These are the objectives it covers, each with practice questions and worked explanations.

Objectives in this domain

Sample question from this domain

Free sampleMultimodal Datamedium

A team fine-tunes an NVIDIA Riva ASR pipeline on domain-specific vocabulary but finds the word error rate remains high on rare technical terms even after acoustic model fine-tuning. Which component should they target next to reduce errors on those terms?

  • AThe vocoder, because it synthesises output waveforms and controls pronunciation fidelity for technical terms.
  • BThe mel spectrogram extractor, because adjusting the filterbank resolution improves how rare phonemes are represented in the feature space.
  • CThe beam-search width parameter, because widening the search ensures more candidate tokens are explored per decoding step.
  • DThe language model component, because it assigns probabilities to word sequences and can be updated with a custom vocabulary or n-gram model. Correct
Distinguish the roles of acoustic and language model components in an ASR pipeline and identify the correct component for vocabulary customisation. NVIDIA Riva's ASR stack separates acoustic modelling (mapping audio features to phonemes or subword units) from language modelling (ranking word-sequence hypotheses). When the acoustic model already captures phoneme patterns correctly but rare technical terms are still misrecognised, the bottleneck is the language model's prior over word sequences. Updating the language model with a custom lexicon or domain-adapted n-gram / neural LM directly improves decoding decisions for those terms, which is reflected as a lower word error rate on the target vocabulary.

Why A is wrong: Vocoders are a TTS component that converts mel spectrograms into audio waveforms; they have no role in ASR decoding or vocabulary coverage.

Why B is wrong: Mel spectrogram extraction is a fixed pre-processing step; changing filterbank parameters affects all phonemes equally and does not address vocabulary coverage gaps.

Why C is wrong: Increasing beam width can marginally improve recall of low-probability paths, but it does not introduce domain vocabulary knowledge and has diminishing returns well before fixing a vocabulary-coverage deficit.

Why D is correct: In a hybrid ASR system the language model scores candidate transcriptions; injecting domain vocabulary and phrase probabilities directly reduces substitution errors on rare terms the acoustic model already produces as phoneme sequences.

Other domains in this exam

See also the NCA-GENM cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by NVIDIA. Original, blueprint-aligned practice material only.