1 |
Exploiting parse structures for native language identification
|
|
|
|
BASE
|
|
Show details
|
|
2 |
Unsupervised syntactic chunking with acoustic cues : computational models for prosodic bootstrapping
|
|
|
|
Abstract:
Learning to group words into phrases without supervision is a hard task for NLP systems, but infants routinely accomplish it. We hypothesize that infants use acoustic cues to prosody, which NLP systems typically ignore. To evaluate the utility of prosodic information for phrase discovery, we present an HMM-based unsupervised chunker that learns from only transcribed words and raw acoustic correlates to prosody. Unlike previous work on unsupervised parsing and chunking, we use neither gold standard part-of-speech tags nor punctuation in the input. Evaluated on the Switchboard corpus, our model outperforms several baselines that exploit either lexical or prosodic information alone, and, despite producing a flat structure, performs competitively with a state-of-the-art unsupervised lexicalized parser, with a substantial advantage in precision. Our results support the hypothesis that acoustic-prosodic cues provide useful evidence about syntactic phrases for language-learning infants. ; 10 page(s)
|
|
Keyword:
080100 Artificial Intelligence and Image Processing
|
|
URL: http://hdl.handle.net/1959.14/301577
|
|
BASE
|
|
Hide details
|
|
4 |
Temporal relation between speech and co-verbal iconic gestures in multimodal interface design
|
|
|
|
BASE
|
|
Show details
|
|
5 |
Collocations in multilingual natural language generation : Lexical functions meet Lexical functional grammar
|
|
|
|
BASE
|
|
Show details
|
|
6 |
Predictability effects in adult-directed and infant-directed speech : does the listener matter?
|
|
|
|
BASE
|
|
Show details
|
|
8 |
Generating subsequent reference in shared visual scenes : computation vs. re-use
|
|
|
|
BASE
|
|
Show details
|
|
10 |
Building an audio-visual corpus of Australian English : large corpus collection with an economical portable and replicable Black Box
|
|
|
|
BASE
|
|
Show details
|
|
11 |
GRE3D7 : A Corpus of distinguishing descriptions for objects in visual scenes
|
|
|
|
BASE
|
|
Show details
|
|
13 |
The Impact of language models and loss functions on repair disfluency detection
|
|
|
|
BASE
|
|
Show details
|
|
14 |
The Impact of visual context on the content of referring expressions
|
|
|
|
BASE
|
|
Show details
|
|
16 |
Producing power-law distributions and damping word frequencies with two-stage language models
|
|
|
|
BASE
|
|
Show details
|
|
17 |
Using ontologies to synchronize change in relational database systems
|
|
|
|
BASE
|
|
Show details
|
|
18 |
Specifying events and their effects in controlled natural language
|
|
|
|
BASE
|
|
Show details
|
|
|
|