Well tuned Large-Vocabulary Continuous Speech Recognition (LVCSR) has been shown to generally be more effective than vocabulary-independent techniques for ranked retrieval of spoken content when one or the other approach is used alone. Tuning LVCSR systems to a topic domain can be costly, however, and the experiments in this paper show that Out-Of-Vocabulary (OOV) query terms can significantly reduce retrieval effectiveness when that tuning is not performed. Further experiments demonstrate, however, that retrieval effectiveness for queries with OOV terms can be substantially improved by combining evidence from LVCSR with additional evidence from vocabulary-independent Ranked Utterance Retrieval (RUR). The combination is performed by using relevance judgments from held-out topics to learn generic (i.e., topic-independent), smooth, non-decreasing transformations from LVCSR and RUR system scores to probabilities of topical relevance. Evaluated using a CLEF collection that includes topics, spontaneous conversational speech audio, and relevance judgments, the system recovers 57% of the mean uninterpolated average precision that could have been obtained through LVCSR domain tuning for very short queries (or 41% for longer queries).

We present a system to index and search conversational speech using a scoring heuristic on the expected posterior counts of phone n-grams in recognition lattices. We report significant improvements in retrieval effectiveness on five human languages over a strong 1-best baseline. The method is shown to improve the utility (mean average precision) of the retrieved lattices' rank order and to do so with a search cost negligible compared to the fastest yet known methods for the linear scanning of phonetic lattices.

This paper describes two new techniques for increasing the accuracy oftopic label assignment to conversational speech from oral history interviews using supervised machine learning in conjunction with automatic speech recognition. The first, time-shifted classification, leverages local sequence information from the order in which the story is told. The second, temporal label weighting, takes the complementary perspective by using the position within an interview to bias label assignment probabilities. These methods, when used in combination, yield between 6% and 15% relative improvements in classification accuracy using a clipped R-precision measure that models the utility of label sets as segment summaries in interactive speech retrieval applications.

We introduce several methods of combining feature selectors for text classification. Results from a large investigation of these combinations are summarized. Easily constructed combinations of feature selectors are shown to improve peak R-precision and F1 at statistically significant levels.

We consider the relationship between training set size and the parameter k for the k-Nearest Neighbors (kNN) classifier. When few examples are available, we observe that accuracy is sensitive to k and that best k tends to increase with training size. We explore the subsequent risk that k tuned on partitions will be suboptimal after aggregation and re-training. This risk is found to be most severe when little data is available. For larger training sizes, accuracy becomes increasingly stable with respect to k and the risk decreases.

No abstract available.