Search

27 Jul 2020: This model uses a long-range history and exploits topic information in the test text to adjust probabilities of test words. ... The PLSA-based model was found to lower test set perplexity over a traditional wordclass-based 4-gram by 13% (optimistic

27 Jul 2020: The focus is primarily on retrieval studies, and on speech tests directly related to retrieval, not on speech recognition itself. ... The report draws on the many and varied tests done during the project, but also presents a new series of results

27 Jul 2020: set. These factors control the point at which training set performance or test set generalisation become the limiting factors. ... The experimental results provide an empirical verification of the training set/test set performance of frame discriminative

27 Jul 2020: all the test data together or 8.6% when processing the test data shows independently. |

27 Jul 2020: The complete system yields an overall word error rate of 22.0% on the 1996 unpartitioned broadcast news development test data and just 15.8% on the 1997 evaluation test set. |

27 Jul 2020: For high dimensional inputs the performance of the single layer network alone is better on the test data, whereas for low dimensional inputs a two layer adaptive network is required and ... this needs much more training before it can produce a better

27 Jul 2020: It consists of speaker-independent (SI) read material, split into training, development test and evaluation test sets.

27 Jul 2020: An initial series of tests are performed in vitro. The statistical theory to predict the increase in signal to noise ratio is developed and verified empirically. ... Having proven the benefits of spatial compounding in vitro, tests with higher levels of

27 Jul 2020: word bigram "hub" tests and the second lowest error rate on the 20k word trigram "hub" test.

27 Jul 2020: word error rate on the 1998 evaluation test set.

27 Jul 2020: These tests have shown the applicability of nonlinear prediction to speech coding and the improvement in coding performance.

27 Jul 2020: In a number of test cases this algorithm is shown to converge reliably to the correct results. |

27 Jul 2020: The final 2003 CU-HTK CTS system constructed from some of these models is described and its performance on the DARPA/NIST 2003 Rich Transcription (RT-03) evaluation test set is

27 Jul 2020: This mismatch' function is then used to estimate the difference in channel conditions between training and test environments.

27 Jul 2020: On the 1998 test the system produced an average word error rate of 16.1% running in 9.5xRT. |

27 Jul 2020: Given insufficient training data, such a model's test set performance will be lower than that of a suitably biased model.

27 Jul 2020: In this paper we extend the HME to classification and results are reported for three common classification benchmark tests: Exclusive-Or, N-input Parity and Two Spirals.

27 Jul 2020: Results are presented for the 2001 development test set and the 2002 evaluation set.

27 Jul 2020: Abstract for niesler_tr265. Cambridge University Engineering Department Technical Report CUED/F-INFENG/TR265. COMPARATIVE EVALUATION OF WORD- AND CATEGORY-BASED LANGUAGE MODELS. Thomas Niesler and Phil Woodland. July 1996. Conventional n-gram

27 Jul 2020: The results show that the proposed method is very much faster yet performs acceptably compared to conventional systems which depend on keyword-specific training or prior knowledge of the test set

27 Jul 2020: An implementation is outlined and demonstrated on test sequences containing two motions.

27 Jul 2020: Because the compensation is word-dependent, its application needs to be hypothesis-driven on the test material.

27 Jul 2020: The paper also presents results on a new set of 60 queries with assessments for the TREC-6 test document data used for development purposes, and analyses the relationship between recognition

27 Jul 2020: It is found that, in most cases, the commonly-used cross-correlation does not perform as well as some other measures, such as the chi squared test or the sum of

27 Jul 2020: Here we address the problem of contaminations (e.g. occlusion, shadows) in test images which have not explicitly been encountered in training data.

27 Jul 2020: This allows successful adaptation on any amount of adaptation data. Tests on the RM1 database show that successful adaptation can be achieved with only 11 seconds of speech, and performance converges

27 Jul 2020: MLLR can also be used in an iterative fashion to refine the transcriptions of the test data and adapt models based on the current transcriptions.

27 Jul 2020: A comparison of test results show that it outperforms one of the best existing methods.

27 Jul 2020: The HTK system yielded the lowest error rates in both the H3-P0 and H3-C0 tests.

27 Jul 2020: Both these schemes require no a-priori thresholding information. An initial evaluation of the techniques was performed using data from the ARPA 1994 test data.

27 Jul 2020: of tests on the LOB corpus to yield very good tagging accuracies.

27 Jul 2020: In this paper, we present a technique to compensate for out-of-plane rotations, which we test using simulations and in vitro experiments.

27 Jul 2020: Front-end and language modelling experiments conducted using various training and test sets from both the Switchboard and Callhome English corpora are presented.

27 Jul 2020: Both techniques yield a significant reduction in perplexity over the baseline trigram language model when faced with multi-domain test text, the mixture-based model giving a 24% reduction and the

27 Jul 2020: Results are presented for the 2001 development and evaluation test sets.

27 Jul 2020: Initial tests show that measurement errors, including image artifacts, are significant enough to obscure and distort diagnostically important information.

27 Jul 2020: Our system achieved a high correct identification rate of 97% on a freely available test set of 29 individuals and extreme illumination changes.

27 Jul 2020: The segmental intelligibility was measured using the Modified Rhyme Test, and a 5.0% error rate obtained.

27 Jul 2020: Tests of Fukushima's original system and the novel systems proposed in this paper suggest that the neocognitron is unable to achieve the rformance of existing digit classifiers due to its

27 Jul 2020: The HTK system yielded the lowest error rates in both the H3-P0 and H3-C0 tests.

27 Jul 2020: We discuss the best way in which to define an optimum architecture, and we then use various ideas from linear algebra to derive an algorithm, which we test experimentally.

27 Jul 2020: The system is evaluated on two test databases, one including speaker-specific X-ray training data and the other acoustic data alone, and improvements in word recognition accuracy are obtained in

27 Jul 2020: Experimental results show that the proposed technique produces good estimates of the human 3D pose on a range of test videos in a cluttered environment.

27 Jul 2020: A technique that optimises the length of individual n-grams is proposed, and experimental tests show it to lead to improved results.

27 Jul 2020: subset; (ii) we include a pose refinement step to optimize the registration between the test image and face exemplar; and (iii) we use robust distance to a sub-space to allow