Semantic Analysis and Organization of Spoken Documents Based on Parameters Derived From Latent Topics

Kong, Sheng-yi; Lee, Lin-Shan

doi:10.1109/tasl.2010.2102592

Cited by 15 publications

(14 citation statements)

References 77 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, speech recognition is the research focus for decades. Kong et al [8] and Wang et al [9] focus on extracting spectro-temporal modulation information for enhancing recognition accuracy. There are research [5,6] focusing on Mandarin Chinese texting input.…”

Section: Related Workmentioning

confidence: 99%

“…Similarly, we use texting to describe the process of composing digital messages encoding with the format for personal computers and mobile devices, typically consisting of alphabetic and numeric characters. Currently, there are three popular input methods including typing, speech recognition [1][2][3][4], and stroke recognition [5][6][7][8][9]. We give the details of each input method in Section 3.…”

Section: Introductionmentioning

confidence: 99%

“…It is even worse on mobile phones due to the fat finger issues. Another popular method is speech recognition [5][6][7][8][9], but recognition rate is problematic, and some environments are not proper for certain environments and situations such as a noisy station or a meeting occasion. We conduct a set of pilot studies to verify and assert these commonly observed issues.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Instant messaging with emotion-embedded vectorized handwritings on mobile devices

Syu

Kuo

Yao

et al. 2017

J Image Video Proc.

View full text Add to dashboard Cite

Generally, handwriting can reflect writers' personality, thoughts, and emotions, i.e., handwriting can deliver emotionand sincereness-embedded messages. However, texting messages and notes such as emails and instant messages replace handwriting letters and notes in communication due to the popularity and availability of mobile devices and personal computers. Furthermore, the commonly used input methods and devices also limit handwriting messaging. For example, limited mobile screen sizes make writing multiple Chinese characters difficult. As a result, this work aims at designing and developing a handwriting messaging system based on our handwriting characteristic exploration and texting-handwriting difference discovery. We first discover Chinese texting issues for mobile devices and emotion-delivered effectiveness of handwriting with pilot studies. Then, our emotion-embedded handwriting messaging framework is implemented to record writing strokes on the entire touch screen, vectorize them as Bézier curves, and send them as instant messages. Vectorizing strokes can preserve personal and emotional handwriting features with compact networking traffic without deteriorating the convenience of instant messaging. Finally, we conduct user studies to verify that our handwriting messaging system is preferred while intending to deliver contents with sincerity and emotion.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Instant messaging with emotion-embedded vectorized handwritings on mobile devices

Syu

Kuo

Yao

et al. 2017

J Image Video Proc.

View full text Add to dashboard Cite

show abstract

“…The PLSA model can be optimized using the EM algorithm, by maximizing a likelihood function [14]. We utilize two parameters from PLSA, latent topic significance (LTS) and latent topic entropy (LTE) [15]. The parameters can also be computed by other topic models, such as latent dirichilet allocation (LDA) [16] in a similar way.…”

Section: Parameters From Topic Modelmentioning

confidence: 99%

“…This measure outperformed the very successful "significance score" [15,10] in speech summarization, so we use the LTE-based statistical measure, s(t i , d), as our baseline.…”

Section: Statistical Measures Of a Termmentioning

confidence: 99%

Two-layer mutually reinforced random walk for improved multi-party meeting summarization

Chen

Metze

2012

2012 IEEE Spoken Language Technology Workshop (SLT)

View full text Add to dashboard Cite

This paper proposes an improved approach of summarization for spoken multi-party interaction, in which a two-layer graph with utterance-to-utterance, speaker-to-speaker, and speakerto-utterance relations is constructed. Each utterance and each speaker are represented as a node in the utterance-layer and speaker-layer of the graph respectively, and the edge between two nodes is weighted by the similarity between the two utterances, the two speakers, or the utterance and the speaker. The relation between utterances is evaluated by lexical similarity via word overlap or topical similarity via probabilistic latent semantic analysis (PLSA). By within-and between-layer propagation in the graph, the scores from different layers can be mutually reinforced so that utterances can automatically share the scores with the utterances from the same speaker and similar utterances. For both ASR output and manual transcripts, experiments confirmed the efficacy of involving speaker information in the two-layer graph for summarization.

show abstract

Semantic Topic Discovery for Lecture Video

Bian

Huang

2019

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

With more and more lecture videos are available on the Internet, on-line learning and e-learning are getting increasing concerns because of many advantages such as high degree of interactivity. The semantic content discovery for lecture video is a key problem. In this paper, we propose a Multi-modal LDA model, which discovers the semantic topics of lecture videos by considering audio and visual information. Specifically, the speaking content and the information of presentation slides are extracted from the lecture videos. With the proposed inference and learning algorithm, the semantic topics of the video can be discovered. The experimental results show that the proposed method can effectively discover the meaningful semantic characters of the lecture videos.

show abstract

Semantic Analysis and Organization of Spoken Documents Based on Parameters Derived From Latent Topics

Cited by 15 publications

References 77 publications

Instant messaging with emotion-embedded vectorized handwritings on mobile devices

Instant messaging with emotion-embedded vectorized handwritings on mobile devices

Two-layer mutually reinforced random walk for improved multi-party meeting summarization

Semantic Topic Discovery for Lecture Video

Contact Info

Product

Resources

About