Automatic Transformation of Lecture Transcription into Document Style using Statistical Framework

Shitaoka, Kazuya; Nanjo, H.; Kawahara, Tatsuya

doi:10.5715/jnlp.11.2_67

Cited by 7 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We consider multiple hypotheses from a decoder by a max operator while traditional methods based on statistical methods consider only the single-best hypothesis and thus use only P(w|s) [7]- [10].…”

Section: Formulationmentioning

confidence: 99%

“…Therefore, to improve the readability of the output text, we need to detect sentence boundaries and recover the periods. While periods do not always correspond to pauses, relationships still exist between them [7]. Therefore, initially we added entries into the translation table as follows:…”

Section: Insertion Of Periodsmentioning

confidence: 99%

“…Operations for correcting transcripts include removing filled pauses and repetition, converting from a spoken style to a written style, inserting punctuation marks such as commas and periods, and discourse markers such as for paragraphs. In this context, there is currently extensive research on paraphrasing and correcting recognition results [7]- [10]. Shitaoka et al formulated the problem as a kind of machine translation and applied a statistical method to transform spoken language to written language [7].…”

Section: Introductionmentioning

confidence: 99%

“…In this context, there is currently extensive research on paraphrasing and correcting recognition results [7]- [10]. Shitaoka et al formulated the problem as a kind of machine translation and applied a statistical method to transform spoken language to written language [7]. Hori et al used weighted finite state transducers (WFSTs) for the same purpose by representing each component as a WFST [8].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Improving the Readability of ASR Results for Lectures Using Multiple Hypotheses and Sentence-Level Knowledge

Fujii¹,

Yamamoto²,

Nakagawa³

2012

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYThis paper presents a novel method for improving the readability of automatic speech recognition (ASR) results for classroom lectures. Because speech in a classroom is spontaneous and contains many ill-formed utterances with various disfluencies, the ASR result should be edited to improve the readability before presenting it to users, by applying some operations such as removing disfluencies, determining sentence boundaries, inserting punctuation marks and repairing dropped words. Owing to the presence of many kinds of domain-dependent words and casual styles, even state-of-the-art recognizers can only achieve a 30-50% word error rate for speech in classroom lectures. Therefore, a method for improving the readability of ASR results is needed to make it robust to recognition errors. We can use multiple hypotheses instead of the single-best hypothesis as a method to achieve a robust response to recognition errors. However, if the multiple hypotheses are represented by a lattice (or a confusion network), it is difficult to utilize sentence-level knowledge, such as chunking and dependency parsing, which are imperative for determining the discourse structure and therefore imperative for improving readability. In this paper, we propose a novel algorithm that infers clean, readable transcripts from spontaneous multiple hypotheses represented by a confusion network while integrating sentence-level knowledge. Automatic and manual evaluations showed that using multiple hypotheses and sentence-level knowledge is effective to improve the readability of ASR results, while preserving the understandability. key words: improving readability of ASR results, confusion network, automatic speech recognition, classroom lecture speech, sentence-level knowledge

show abstract

Section: Formulationmentioning

confidence: 99%

Section: Insertion Of Periodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Improving the Readability of ASR Results for Lectures Using Multiple Hypotheses and Sentence-Level Knowledge

Fujii¹,

Yamamoto²,

Nakagawa³

2012

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Shitaoka et al [4] presented a noisy-channel model for speaking style transformation (SST). We expanded this model through a weighted finite state transducer (WFST) implementation and the introduction of a variety of features in a log-linear framework [5].…”

Section: Introductionmentioning

confidence: 99%

Improved statistical models for SMT-based speaking style transformation

Neubig

Akita

Mori

et al. 2010

2010 IEEE International Conference on Acoustics, Speech and Signal Processing

Self Cite

View full text Add to dashboard Cite

Automatic speech recognition (ASR) results contain not only ASR errors, but also disfluencies and colloquial expressions that must be corrected to create readable transcripts. We take the approach of statistical machine translation (SMT) to "translate" from ASR results into transcript-style text. We introduce two novel modeling techniques in this framework: a context-dependent translation model, which allows for usage of context to accurately model translation probabilities, and log-linear interpolation of conditional and joint probabilities, which allows for frequently observed translation patterns to be given higher priority. The system is implemented using weighted finite state transducers (WFST). On an evaluation using ASR results and manual transcripts of meetings of the Japanese Diet (national congress), the proposed methods showed a significant increase in accuracy over traditional modeling techniques.

show abstract

A study on output sentence generation method for question answering using statistical machine translation

Yamada

Arakawa

2013

2013 13th International Conference on Control, Automation and Systems (ICCAS 2013)

View full text Add to dashboard Cite

Question answering generally generates the answer to the question by extracting the named entity from the sentences containing the answer to the question from information sources. However, it is not always true that a named entity is an answer to the question. So we propose a method for generating the answer sentence using statistical machine translation. The probability models are constructed by learning from enormous samples of the set of question sentence, extracted sentence, and answer sentence. The question sentence and the sentence extracted by the question answering from information source are regarded as an input of machine translation. They are translated to a suitable answer sentence to the question. In this paper, we attempted to apply our method to several simple types questions that can also be answered by the named entity extraction.

show abstract

Automatic Transformation of Lecture Transcription into Document Style using Statistical Framework

Cited by 7 publications

References 0 publications

Improving the Readability of ASR Results for Lectures Using Multiple Hypotheses and Sentence-Level Knowledge

Improving the Readability of ASR Results for Lectures Using Multiple Hypotheses and Sentence-Level Knowledge

Improved statistical models for SMT-based speaking style transformation

A study on output sentence generation method for question answering using statistical machine translation

Contact Info

Product

Resources

About