Linear Discourse Segmentation of Multi-Party Meetings Based on Local and Global Information

Bokaei, Mohammad Hadi; Sameti, Hossein; Liu, Yang

doi:10.1109/taslp.2015.2456430

Cited by 4 publications

(9 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Annotation and machine learning of discourse segmentation covers several types of units, including topical segments (Galley et al, 2003), meeting units in which action items are identified or decisions made (Purver et al, 2007;Fernández et al, 2008), transaction subtasks for ordering library books , or speaker involvement (Bokaei et al, 2015). This work relies on manual transcription, and draws on many sources of knowledge for machine learned models, including turn-taking, prosody, and linguistic features.…”

Section: Resultsmentioning

confidence: 99%

“…This work relies on manual transcription, and draws on many sources of knowledge for machine learned models, including turn-taking, prosody, and linguistic features. The segmentation annotation can be linear (Galley et al, 2003;Bokaei et al, 2015;Passonneau and Litman, 1997; or hierarchical (Purver et al, 2007;Fernández et al, 2008;Passonneau et al, 2011). The differences in methods and results across this body of work, points to a need for more datasets for research on the organization of discourse into activity units.…”

Section: Resultsmentioning

confidence: 99%

“…In (Galley et al, 2003), three annotators segmented the ICSI meeting corpus into topical units, and majority agreement was taken as ground truth. A functional segmentation of meetings from the AMI multiparty meeting corpus based on involved participants was segmented by one annotator and finalized by a second annotator (Bokaei et al, 2015). Task-based segmentation of patron-librarian interactions (Passonneau et al, 2011) measured agreement among two annotators using Krippendorff's Alpha at an average of 0.77 (Krippendorff, 1980).…”

Section: Related Workmentioning

confidence: 99%

“…Speakers shift from one activity to the next with more or less awareness and explicit demarcation. To treat such conversational activities as a sequence of discrete units is a convenient oversimplification that is often resorted to (Bokaei et al, 2015;Galley et al, 2003;Passonneau and Litman, 1997). Systems that provide automated access to spoken language data often rely on segmentation of spoken discourse into sequential units for summarization (Wang and Cardie, 2012;Dielmann and Renals, 2005) or information retrieval (Ward et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Estimation of Discourse Segmentation Labels from Crowd Data

Huang

Zhong

Passonneau³

2015

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

For annotation tasks involving independent judgments, probabilistic models have been used to infer ground truth labels from data where a crowd of many annotators labels the same items. Such models have been shown to produce results superior to taking the majority vote, but have not been applied to sequential data. We present two methods to infer ground truth labels from sequential annotations where we assume judgments are not independent, based on the observation that an annotator's segments all tend to be several utterances long. The data consists of crowd labels for annotation of discourse segment boundaries. The new methods extend Hidden Markov Models to relax the independence assumption. The two methods are distinct, so positive labels proposed by both are taken to be ground truth. In addition, results of the models are checked using metrics that test whether an annotator's accuracy relative to a given model remains consistent across different conversations.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Estimation of Discourse Segmentation Labels from Crowd Data

Huang

Zhong

Passonneau³

2015

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

show abstract

“…Boakye and Trueba-Hornero focused in their respective theses [110,111] and publications [112,113] on developing an overlap detection system for monaural recordings. The detection framework was relying on an eHMM segmenter, which segmented the signal into overlap, non-overlap, and non-speech class.…”

Section: Single Distant Microphone Methodsmentioning

confidence: 99%

Detection and handling of overlapping speech for speaker diarization

Zelenák

View full text Add to dashboard Cite

For the last several years, speaker diarization has been attracting substantial research attention as one of the spoken language technologies applied for the improvement, or enrichment, of recording transcriptions. Recordings of meetings, compared to other domains, exhibit an increased complexity due to the spontaneity of speech, reverberation effects, and also due to the presence of overlapping speech. Overlapping speech refers to situations when two or more speakers are speaking simultaneously. In meeting data, a substantial portion of errors of the conventional speaker diarization systems can be ascribed to speaker overlaps, since usually only one speaker label is assigned per segment. Furthermore, simultaneous speech included in training data can eventually lead to corrupt single-speaker models and thus to a worse segmentation. This thesis concerns the detection of overlapping speech segments and its further application for the improvement of speaker diarization performance. We propose the use of three spatial cross-correlationbased parameters for overlap detection on distant microphone channel data. Spatial features from different microphone pairs are fused by means of principal component analysis, linear discriminant analysis, or by a multi-layer perceptron. In addition, we also investigate the possibility of employing longterm prosodic information. The most suitable subset from a set of candidate prosodic features is determined in two steps. Firstly, a ranking according to mRMR criterion is obtained, and then, a standard hill-climbing wrapper approach is applied in order to determine the optimal number of features. The novel spatial as well as prosodic parameters are used in combination with spectral-based features suggested previously in the literature. In experiments conducted on AMI meeting data, we show that the newly proposed features do contribute to the detection of overlapping speech, especially on data originating from a single recording site. In speaker diarization, for segments including detected speaker overlap, a second speaker label is picked, and such segments are also discarded from the model training. The proposed overlap labeling technique is integrated in Viterbi decoding, a part of the diarization algorithm. During the system development it was discovered that it is favorable to do an independent optimization of overlap exclusion and labeling with respect to the overlap detection system. We report improvements over the baseline diarization system on both single- and multi-site AMI data. Preliminary experiments with NIST RT data show DER improvement on the RT ¿09 meeting recordings as well. The addition of beamforming and TDOA feature stream into the baseline diarization system, which was aimed at improving the clustering process, results in a bit higher effectiveness of the overlap labeling algorithm. A more detailed analysis on the overlap exclusion behavior reveals big improvement contrasts between individual meeting recordings as well as between various settings of the overlap detection operation point. However, a high performance variability across different recordings is also typical of the baseline diarization system, without any overlap handling.

show abstract