Article Structure
Abstract
We introduce a novel mechanism for incorporating articulatory dynamics into speech recognition with the theory of task dynamics.
Introduction
Although modern automatic speech recognition (ASR) takes several cues from the biological perception of speech, it rarely models its biological production.
Background and related work
The use of theoretical (phonological) features of the vocal tract has provided some improvement over traditional acoustic ASR systems in phoneme recognition with neural networks (Kirchhoff, 1999; Roweis, 1999), but there has been very little work in ASR informed by direct measurements of the vocal tract.
Baseline systems
We now turn to the task of speech recognition.
Switching Kalman filter
Our first experimental system attempts speech recognition given only articulatory data.
Recognition with task dynamics
Our goal is to integrate task dynamics within an ASR system for continuous sentences called TD-ASR.
Experiments
Experimental data is obtained from two sources, as described in section 2.2.
Discussion and conclusions
The articulatory medium of speech rarely informs modern speech recognition.
Topics
baseline systems
- Experiments compare this with two baseline systems , namely an acoustic hidden Markov model and a dynamic Bayes network augmented with discretized representations of the vocal tract.Page 1, “Abstract”
- We examine two baseline systems .Page 3, “Baseline systems”
- Figure 3: Baseline systems : (a) acoustic hidden Markov model and (b) articulatory dynamic Bayes network.Page 4, “Baseline systems”
- For each of our baseline systems , we calculate the phoneme-error—rate (PER) and word-error-rate (WER) after training.Page 6, “Experiments”
- Table 1: Phoneme- and Word-Error-Rate (PER and WER) for different parameterizations of the baseline systems .Page 6, “Experiments”
See all papers in Proc. ACL 2010 that mention baseline systems.
See all papers in Proc. ACL that mention baseline systems.
Back to top.
