Ikibazo gihuza ibibazo · Biragoye Urwego
WFST Kumenyekanisha Ikibazo
Kurikirana ibihimbano bya HCLG hamwe nuburemere bwa WFST, uhuze ibice bishushanyo ninkomoko yubumenyi bwo kuvuga, kandi utandukanye imvange na decoding ya nyuma.
Ikibazo 1 Bya 8
Mubisanzwe bisanzwe bya HCLG, L igereranya iki?
Reba buri kibazo8 ibibazo
Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.
- Mubisanzwe bisanzwe bya HCLG, L igereranya iki?
- The acoustic waveform
- The context-dependency transducer
- The pronunciation lexicon mapping phones and words
- The grammar or language model
- Which component encodes word-sequence grammar or language-model constraints in HCLG?
- H
- G
- C
- L
- Niki arc muri uburemere buremereye bwa leta transducer ikunze gutwara?
- State transitions and weights only, with no label mapping
- Input labels only, without output labels or weights
- Word probabilities without graph states
- Input-output labels and weights
- Guhimba transducers bihuye niki?
- Combines sequential mappings into a composed search graph
- Adds graph arcs without checking whether symbol labels match
- Discards output labels and keeps only acoustic states
- Chooses the final recognized words without using acoustic observations
- Niki moderi ya acoustique itanga mugihe cyo kumenyekanisha imvange?
- Only the final grammar
- The text normalization dictionary
- Scores associated with acoustic states or transitions
- The lexicon's word spellings
- Kuki ikibonezamvugo kigufi gishobora guteza akaga muri decoder?
- It guarantees too many pronunciations
- It may exclude valid expressions outside its allowed paths
- It removes the acoustic model
- It makes symbol tables unnecessary
- What must align for WFST composition to work correctly?
- Every component has an identical number of states
- All components use numerically identical weight scales
- Every graph arc has the same cost
- Compatible symbol labels between component transducers
- Which claim is appropriately scoped to HCLG?
- It is a graph composition used in conventional hybrid HMM recipes
- It describes every end-to-end neural decoder
- It is the name of a language model alone
- It guarantees perfect transcription