Spice+: Evaluation of Automatic Audio Captioning Systems with Pre-Trained Language Models.
Félix GontierRomain SerizelChristophe CerisaraPublished in: ICASSP (2023)
Keyphrases
- language model
- speech recognition
- pre trained
- language modeling
- document retrieval
- n gram
- probabilistic model
- language modelling
- test collection
- information retrieval
- document ranking
- retrieval model
- context sensitive
- relevance assessments
- multimedia
- smoothing methods
- query expansion
- retrieval systems
- hidden markov models
- prior knowledge