Tone Modeling for Continuous Mandarin Speech Recognition

Cao, Yang; Zhang, Shuwu; Huang, Taiyi; Xu, Bo

doi:10.1023/b:ijst.0000017012.11970.6a

Cited by 9 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this data, there were 178,199 Chinese words and 183,452 English words; Chinese vocabulary size was 8,687, while English vocabulary size was 6,956. SR output-lattices and n-best lists-was generated for Mandarin using software developed by the Institute of Automation, Chinese Academy of Sciences (Cao et al 2004;Ding and Xu 2004). The 1-best word error rate was 0.420; the 1-best sentence error rate was 0.806; and the 20-best word error rate was 0.253.…”

Section: Slt System Components and User Cooperationmentioning

confidence: 99%

Toward Practical Spoken Language Translation

Zong

Seligman

2005

Machine Translation

View full text Add to dashboard Cite

This paper argues that the time is now right to field practical Spoken Language Translation (SLT) systems. Several sorts of practical systems can be built over the next few years if system builders recognize that, at the present state of the art, users must cooperate and compromise with the programs. Further, SLT systems can be arranged on a scale, in terms of the degree of cooperation or compromise they require from users. In general, the broader the intended linguistic or topical coverage of a system, the more user cooperation or compromise it will presently require. The paper briefly discusses the component technologies of SLT systems as they relate to user cooperation and accommodation ("human factors engineering"), with examples from the authors' work. It describes three classes of "cooperative" SLT systems which could be put into practical use during the next few years.

show abstract

Section: Slt System Components and User Cooperationmentioning

confidence: 99%

Toward Practical Spoken Language Translation

Zong

Seligman

2005

Machine Translation

View full text Add to dashboard Cite

show abstract

“…In addition to these four tones, there's also a neutral tone, which itself doesn't possess a particular pitch contour pattern. It is essentially a short and unstressed continuation of the previous syllable, so its pitch contour is relatively arbitrary [1,6,7]. This type is denoted as T5.…”

Section: Introduction 1backgroundmentioning

confidence: 99%

End-to-End Mandarin Tone Classification with Short Term Context Information

Tang¹,

Li²

2021

Preprint

View full text Add to dashboard Cite

In this paper, we propose an end-to-end Mandarin tone classification method from continuous speech utterances utilizing both the spectrogram and the short term context information as the inputs. Both Mel-spectrograms and context segment features are used to train the tone classifier. We first divide the spectrogram frames into syllable segments using force alignment results produced by an ASR model. Then we extract the short term segment features to capture the context information across multiple syllables. Feeding both the Mel-spectrogram and the short term context segment features into an end-to-end model could significantly improve the performance. Experiments are performed on a large scale open source Mandarin speech dataset to evaluate the proposed method. Results show that the this method improves the classification accuracy from 79.5% to 88.7% on the AISHELL3 database.

show abstract

“…Continuous speech recognition (CSR) for tonal languages have been extensively studied in languages like Cantonese, Mandarin, Thai, Vietnamese. Previous studies have reported that tones carry important lexical information in tonal languages which serves as an important cue for speech recognition tasks [1][2][3][4][5][6][7]. In Cantonese speech recognition task, 8% relative improvement in character error rate is achieved using weighted tone information [2].…”

Section: Introductionmentioning

confidence: 99%

“…In Cantonese speech recognition task, 8% relative improvement in character error rate is achieved using weighted tone information [2]. Cao et al reported that tone information integration helped in achieving a 16.2% relative character error rate reduction in continuous Mandarin speech recognition [4]. Similarly, in Mandarin Broadcast News speech recognition task, Lei et al [5] reported the reduction of the character error rate from 13.0% to 11.5% on CTV test set.…”

Section: Introductionmentioning

confidence: 99%

Robust Mizo Continuous Speech Recognition

et al. 2018

View full text Add to dashboard Cite

Mizo is an under-resourced tonal language that is mainly spoken in NorthEast India. It has 4 canonical tones along with a tone-sandhi. In Mizo language, a majority of the words contain tone information. As a result of that, it exhibits higher acoustic variability like other tonal languages in the world. In this work, we investigate the impact of tonal information on robust Mizo continuous speech recognition (CSR). First, separate baseline CSR systems are developed employing the Mel-frequency cepstral coefficient (MFCC) based acoustic features and salient acoustic modeling paradigms. For further improvement, the tonal information has been incorporated in each of the CSR systems. For this purpose, 3-dimensional tonal features are derived which include pitch, pitch-difference, and probability of voicing values. Our experimental study reveals that with the inclusion of tonal information, the robustness of Mizo CSR system gets enhanced across all acoustic modeling paradigms. This trend is attributed to lesser degradation in the fundamental frequency information than the vocal tract information under noisy conditions.

show abstract

Tone Modeling for Continuous Mandarin Speech Recognition

Cited by 9 publications

References 14 publications

Toward Practical Spoken Language Translation

Toward Practical Spoken Language Translation

End-to-End Mandarin Tone Classification with Short Term Context Information

Robust Mizo Continuous Speech Recognition

Contact Info

Product

Resources

About