Crowd-sourcing is a promising method for fast and cheap transcription of large volumes of speech data. However, this method cannot achieve the accuracy of expert transcribers on speech that is difficult to transcribe. Faced with such speech data, we developed three new methods of crowd-sourcing, which allow explicit trade-offs among precision, recall, and cost. The methods are: incremental redundancy, treating ASR as a transcriber, and using a regression model to predict transcription reliability. Even though the accuracy of individual crowd-workers is only 55% on our data, our best method achieves 90% accuracy on 93% of the utterances, using only 1.3 crowd-worker transcriptions per utterance on average. When forced to transcribe all utterances, our best method matches the accuracy of previous crowd-sourcing methods using only one third as many transcriptions. We also study the effects of various task design factors on transcription latency and accuracy, some of which have not been reported before.

As large-scale deployments of spoken dialog systems in call centers become more common, a wealth of information is gathered about the call center business as well as the operation of these systems from their daily logs. This paper describes the "VoiceTone Daily News" data mining tool for analyzing this information and presenting it in a readily comprehensible and customizable form that is suitable for use by anyone from system designers to call center businesses. Relevant business and dialog features are extracted from the speech logs of caller-system interactions and tracked by trend analysis algorithms. We describe novel techniques for generating alerts on multiple data streams while avoiding redundant "knock-on" alerts. Some initial experiments with automated measures of dialog success are described as possible additional features to track. Features that move outside their expected bounds on a given day generate headlines as part of a website generated completely automatically from each day's logs. A "drill-down" facility allows headlines to be investigated all the way to viewing logs of individual interactions behind the headline and listening to the audio for individual turns.

This paper is concerned with language modeling (LM) for large vocabulary speech recognition in Mandarin Chinese. As the language characteristics of Chinese are quite unique, we investigate some novel techniques in language modeling. We also borrow some of techniques that have been applied to other languages. Experiments have been conducted on the Call Home Mandarin, HUB4, and HUB5 corpora obtained from the Linguistic Data Consortium (LDC). The training set consists of 9.8 hours of spontaneous speech and 100K words in text. The test set consists of 1.6 hours of spontaneous speech and 20K words in text. We have found that our results compare favorably to the results reported in the literature.

We present a new motion compensated DCT based video coding scheme. The quantization of DCT coefficients is done using trellis based-scalar vector quantizer (TB-SVQ) introduced by Laroia and Farvardin (see IEEE Transactions on Information Theory, vol.39, no.5, p.851-868, 1993). TB-SVQ is fixed rate and for memoryless sources is known to asymptotically achieve all gains over uniform scalar quantizers. Our coding scheme does not require buffering or variable length coding to reach its target bit rate. Thus the resulting coding delay is minimal, while the performance is very good. Furthermore since no variable length coding is necessary, our coder is robust in the presence of channel errors. Hence it is suitable for video teleconferencing.>