Hubert Jin

Probabilistic models for topic detection and tracking

We present probabilistic models for use in detecting and tracking topics in broadcast news stories. Our information retrieval (IR) models are formally explained. The topic detection and tracking (TDT) initiative is discussed. The application of probabilistic models to the topic detection and tracking tasks is developed, and enhancements are discussed. We discuss four variations of these models, and we report our preliminary test results from the current TDT corpus.

Transcribing radio news

We have recently extended the capabilities of BBNs large-vocabulary discrete-utterance speech recognition system (BYBLOS) to operate on raw audio recordings of radio news programming. The recordings are given to the system as large monolithic waveforms without any additional side-information. Our goal is to transcribe all speech in the input with the highest accuracy possible. The problem is very challenging because radio news programming has frequent changes in speaker, speaking style, dialect, accent, topic, channel and environmental conditions. Furthermore, the monolithic input presents new problems for recognition algorithms and language models since all useful boundaries (such as speaker turns or sentence ends) are unknown.

Broadcast news transcription

We describe our work on automatic transcription of radio and television news broadcasts. This problem is very challenging for large vocabulary speech recognition because of the frequent and unpredictable changes that occur in speaker, speaking style, topic, channel, and background conditions. Faced with such a problem, there is a strong tendency to try to carve the input into separable classes and deal with each one independently. In our early work on this problem, however, we are finding that the rewards for condition-specific techniques are disappointingly small. This is forcing us to look for general, robust, and adaptive algorithms for dealing with extremely variable data. We describe the BBN BYB-LOS recognition system configured to handle off-line transcription and we characterize the speech contained in the 1996 DARPA Hub-4 testbed. On the partitioned development test set, we achieved a 29.4% overall word error rate.

Fast robust inverse transform speaker adapted training using diagonal transformations

We present a new method of speaker adapted training (SAT) that is more robust, faster, and results in lower error rate than the previous methods. The method, called inverse transform SAT (IT-SAT) is based on removing the differences between speakers before training, rather than modeling the differences during training. We develop several methods to avoid the problems associated with inverting the transformation. In one method, we interpolate the transformation matrix with an identity or diagonal transformation. We also apply constraints to the matrix to avoid estimation problems. Finally, we show that the resulting method is much faster, requires much less disk space, and results in higher accuracy than the original SAT method.