Analysis and processing of lecture audio data

Glass, James; Hazen, Timothy J.; Hetherington, Lee; Wang, Chao

doi:10.3115/1626307.1626309

Cited by 70 publications

(51 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We conducted spoken utterance retrieval experiments with lecture speech recordings in MIT lecture corpus [5]. The corpus is a collection of audio-visual recordings of lectures and seminars presented at MIT, which is approximately 300 hours containing lectures from eight different courses and from 80 seminars given on a variety of topics.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Open-Vocabulary Spoken Utterance Retrieval using Confusion Networks

Hori¹,

Hetherington²,

Hazen³

et al. 2007

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07

Self Cite

View full text Add to dashboard Cite

This paper presents a novel approach to open-vocabulary spoken utterance retrieval using confusion networks. If out-of-vocabulary (OOV) words are present in queries and the corpus, word-based indexing will not be sufficient. For this problem, we apply phone confusion networks and combine them with word confusion networks. With this approach, we can generate a more compact index table that enables robust keyword matching compared with typical lattice-based methods. In the retrieval experiments with speech recordings in MIT lecture corpus, our method using phone confusion networks outperformed lattice-based methods especially for OOV queries.

show abstract

Section: Methodsmentioning

confidence: 99%

“…In SUR experiments with lecture speech recordings from MIT lecture corpus [5], we compared several indexing methods including word/phone confusion networks and the combined networks.…”

Section: Introductionmentioning

confidence: 99%

Open-Vocabulary Spoken Utterance Retrieval using Confusion Networks

Hori¹,

Hetherington²,

Hazen³

et al. 2007

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07

Self Cite

View full text Add to dashboard Cite

show abstract

“…Lecture transcription has been the target of much bigger research projects such as the Japanese project described in [2], the European project CHIL (Computers In The Human Communication Loop) [3], and the American iCampus Spoken Lecture Processing project [4]. In some of these projects, the concept of lecture is different.…”

Section: Introductionmentioning

confidence: 99%

Classroom Lecture Recognition

Trancoso

Nunes

Neves

2006

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The main goal of this work is to provide automatic transcriptions of classroom lectures for e-learning and e-inclusion applications. The first experiments using a recognition system trained for Broadcast News resulted in word error rates near 60%, clearly confirming the need for adaptation to the specific topic of the lectures, on one hand, and for better strategies for handling spontaneous speech. This paper describes the different domain adaptation steps that lowered the error rate to 45%, with very little transcribed adaptation material. It also includes a qualitative analysis of the different types of error, focusing on the ones related to a very high rate of disfluencies.

show abstract

“…The speech data used in this paper is taken from a corpus of audio lectures collected at MIT [1]. The entire corpus consists of approximately 300 hours of lectures from a variety of academic courses and seminars.…”

Section: Speech Datamentioning

confidence: 99%

“…In particular, we have observed that for many educational lectures, the active vocabulary is typically very small, but includes Support for this research was provided in part by the National Science Foundation under grant #IIS-0415865. many topic specific terms and phrases that are not frequently occurring in conversational speech [1]. Recognition experiments on these lectures show that reducing the OOV rate improves accuracy, but that including unnecessary words is detrimental to performance [2].…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Word Acquisition from Speech using Pattern Discovery

Park

Glass

2006 IEEE International Conference on Acoustics Speed and Signal Processing Proceedings

View full text Add to dashboard Cite

In this paper, we present an unsupervised method for automatically discovering words from speech using a combination of acoustic pattern discovery, graph clustering, and baseform searching. The algorithm we propose represents an alternative to traditional methods of speech recognition and makes use of the acoustic similarity of multiple realizations of the same words or phrases. On a set of three academic lectures on different subjects, we show that the clustering component of the algorithm is able to successfully generate word clusters that have good coverage of subject-relevant words. Moreover, we illustrate how to use the cluster nodes to retrieve the word identity of each cluster from a large baseform dictionary. Results indicate that this algorithm may prove useful for applications such as vocabulary initialization, speech summarization, or augmentation of existing recognition systems.

show abstract

Analysis and processing of lecture audio data

Cited by 70 publications

References 14 publications

Open-Vocabulary Spoken Utterance Retrieval using Confusion Networks

Open-Vocabulary Spoken Utterance Retrieval using Confusion Networks

Classroom Lecture Recognition

Unsupervised Word Acquisition from Speech using Pattern Discovery

Contact Info

Product

Resources

About