Browsing by Author "Wang, YR"

Now showing 1 - 11 of 11

A high-performance Min-Nan/Taiwanese TTS system
(2003)
Kuo, WC
;
Zhong, XR
;
Wang, YR
;
Chen, SH
;
電信工程研究所
In this paper, the implementation of a high-performance Min-Nan/Taiwanese TTS system is presented. The system can convert both Min-Nan/Taiwanese texts, represented in a hybrid Han-Lo written form, and Chinese texts into natural Taiwanese speeches. It is an improved version of the system developed previously [11]. Improvements include: the add of a "Chinese-to-Min-Nan/Taiwanese" lexicon to solve the OOV problem and to increase the ability of processing Chinese text; the use of explicit tone sandhi rules to ease the learning of prosody generation; a further processing of the training database to detect all breaks not associated with PMs; and the use of four RNNs to separately generate four types of prosodic parameters. The system is implemented by software and runs in real-time on PC. An informal subjective listening test confirmed that the system performed well. All synthetic speeches sounded natural for well-tokenized Min-Nan/Taiwanese texts and for automatic tokenized Chinese texts.
A Mandarin text-to-speech system
(1996)
Hwang, SH
;
Chen, SH
;
Wang, YR
;
交大名義發表
;
電信工程研究所
In this paper, the implementation of a high-performance Mandarin TTS system is presented. The system is composed of four main parts: text analysis (TA), prosodic information generation (PIG), waveform table (WT) of 411 base-syllables, and PSOLA-based waveform synthesis (PSOLA). In TA, a statistical model based method is first employed to automatically tag the input text to obtain the word sequence and the associated part-of-speech (POS) sequence. A lexicon containing about 80000 words is used in the tagging process. Then the corresponding base-syllable sequence is found and used to get from WT the basic wave-form sequence. Some linguistic features used in PIG are also extracted in TA, In PIG, a four-layer recurrent neural network (RNN) is employed to generate some prosodic information including pitch. contour, energy level, initial duration and final duration of syllable as well as inter-syllable pause duration. Finally, in PSOLA the basic waveform sequence is modified using the prosodic information to generate output synthetic speech, The whole system is implemented by software on a PC/AT 486 with a 16-bit Sound Blaster add-on card. Only 3.2 Mbyte memory space is required. It can synthesize speech in real-time for any input Chinese text. Informal listening tests by many native Chinese living in Taiwan confirmed that the synthetic speech sounded very fluent and natural.
A model-based tone labeling method for Min-Nan/Taiwanese speech
(2004)
Kuo, WC
;
Wang, YR
;
Chen, SH
;
電信工程研究所
;
Institute of Communications Engineering
in this paper, a model-based tone labeling method for Min-Nan/Taiwanese speech is proposed. It takes the mean and shape of syllable pitch contour as two modeling units and considers some major affecting factors that control their variations. By using the EM algorithm to estimate all parameters of the pitch mean and shape models from a speech database, we can decide the best tone sequences pronounced in all utterance of the database. Experimental results showed that it outperformed the VQ classification method which suffers from the interferences resulted from neighboring syllables and from the global prosodic phrase pattern.
A new common component GMM-based speaker recognition method
(2005)
Wang, YR
;
Chiang, CY
;
電信工程研究所
;
Institute of Communications Engineering
In this paper, a new common component GMM (CCGMM)based speaker recognition approach is presented. It first defines a divergence measure to calculate the similarity of the speech signals of two speakers. Then, a CCGMM training algorithm which simultaneously maximizes the likelihood of CCGMM and the inter-speaker divergence is proposed. Performance of the proposed approach was examined using a telephone-speech database (MAT) containing 2962 speakers. A speaker recognition rate of 90.0% was achieved. The recognition rate raised to 96.1% when it was combined with the conventional GMM-based scheme.
A new duration modeling approach for Mandarin speech
(2003-07-01)
Chen, SH
;
Lai, WH
;
Wang, YR
;
電信工程研究所
;
Institute of Communications Engineering
In this paper, a new duration modeling approach for Mandarin speech is proposed. It explicitly takes several major affecting factors as multiplicative companding factors (CFs) and estimates all model parameters by an EM algorithm. Besides, the three basic Tone 3 patterns (i.e., full tone, half tone and sandhi tone) are also properly considered via using three different Us to separate their affections on syllable duration. Experimental results showed that the variance of the syllable duration was greatly reduced from 180.17 to 2.52 frame(2) (1 frame =5 ms) by the syllable duration modeling to eliminate effects from those affecting factors. Moreover, the estimated Us of those affecting factors agreed well to our prior linguistic knowledge. Two extensions of the duration modeling method are also performed. One is the use of the same technique to model initial and final durations. The other is to replace the multiplicative model with an additive one. Lastly, a preliminary study of applying the proposed model to predict syllable duration for TTS is also performed. Experimental results showed that it outperformed the conventional regressive prediction method.
A prototype of Mandarin speech telephone number inquiry system
(1997)
Lu, PR
;
Hong, WT
;
Chiang, SL
;
Wang, YR
;
Chen, SH
In this paper, a prototype of Mandarin speech telephone number inquiry system is implemented on a PC platform with a Dialogic D/41D card and a 16-bit Soundblaster card as the telephone line interface. It is an integration of a speaker-independent, isolated-word speech recognition sub-system and a text-to-speech (TTS) subsystem, and is operating under Windows 95. The vocabulary to be recognized comprises 1922 names of bank and insurance companies with word lengths ranging from 2 to 14 syllables. A recognition rate of 92.6% was achieved. All output speech messages are responded by the TTS sub-system.
A statistical pitch detection algorithm
(2002)
Wang, YR
;
Wong, IJ
;
Tsao, TC
;
電信工程研究所
;
Institute of Communications Engineering
A new statistical pitch detection algorithm is proposed in this paper. It first defines frame-based voiced/unvoiced probabilistic measures and between-frame pitch transition probabilities, and then tightly integrates them into an ML search to find the best pitch contour of an utterance. Compared with the pitch detection method of the well-known ESPS software package, the proposed method performs better on both UN decision and pitch frequency detection accuracy.
A statistics-based pitch contour model for Mandarin speech
(2005-02-01)
Chen, SH
;
Lai, WH
;
Wang, YR
;
電信工程研究所
;
Institute of Communications Engineering
A statistics-based syllable pitch contour model for Mandarin speech is proposed. This approach takes the mean and the shape of a syllable log-pitch contour as two basic modeling units and considers several affecting factors that contribute to their variations. The affecting factors include the speaker, prosodic state (which essentially represents the high-level linguistic components of F0 and will be explained more clearly in Sec. 1), tone, and initial and final syllable classes. The parameters of the two modeling units were automatically estimated using the expectation-maximization (EM) algorithm. Experimental results showed that the root mean squared errors (RMSEs) obtained in the closed and open tests in the reconstructed pitch period were 0.362 and 0.373 ins, respectively. This model provides a way to separate the effects of several major factors. All of the inferred values of the affecting factors were in close agreement with our prior linguistic knowledge. It also gives a quantitative and more complete description of the coarticulation effect of neighboring tones rather than conventional qualitative descriptions of the tone sandhi rules. In addition, the model can provide useful cues to determine the prosodic phrase boundaries, including those occurring at intersyllable locations, with or without punctuation marks. (C) 2005 Acoustical Society of America.
A study on Mandarin broadcast news speech recognition
(2004)
Chen, CL
;
Wang, YR
;
Chen, SH
;
電信工程研究所
;
Institute of Communications Engineering
In this paper, a basic Mandarin broadcast news speech recognition system is constructed using the MATBN database. It considers the acoustic modeling for Mandarin base-syllables, particles, and paralinguistic phenomena. It also considers environment-dependent acoustic modeling for three recording environments: studio anchors, outdoor reporters, and outdoor interviewee. Moreover, it incorporates a bigram language model with adaptation using data in MATBN. Syllable recognition rates of 89.64, 84.42 and 61.62% were achieved for the three environments of anchors, reporters and interviewees, respectively.
An RNN-based prosodic information synthesizer for Mandarin text-to-speech
(1998-05-01)
Chen, SH
;
Hwang, SH
;
Wang, YR
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
A new RNN-based prosodic information synthesizer for Mandarin Chinese text-to-speech (TTS) is proposed in this paper. Its four-layer recurrent neural network (RNN) generates prosodic information such as syllable pitch contours, syllable energy levels, syllable initial and final durations, as well as inter-syllable pause durations. The input layer and first hidden layer operate with a word-synchronized clock to represent current-word phonologic states within the prosodic structure of text to be synthesized. The second hidden layer and output layer operate on a syllable-synchronized clock and use outputs from the preceding layers, along with additional syllable-level inputs fed directly to the second hidden layer, to generate desired prosodic parameters. The RNN was trained on a large set of actual utterances accompanied by associated texts, and can automatically learn many human-prosody phonologic rules, including the well-known Sandhi Tone 3 F0-change rule. Experimental results show that all synthesized prosodic parameter sequences matched quite well with their original counterparts, and a pitch-synchronous-overlap-add-based (PSOLA-based) Mandarin TTS system was also used for testing of our approach. While subjective tests are difficult to perform and remain to be done in the future, we have carried out informal listening tests by a significant number of native Chinese speakers and the results confirmed that all synthesized speech sounded quite natural.
Mandarin telephone: Speech recognition for automatic telephone number directory service
(1998)
Wang, YR
;
Chen, SH
;
電信工程研究所
;
Institute of Communications Engineering
This paper discusses an HMM-based Mandarin telephone speech recognition method for implementing a prototype system of automatic telephone number directory service. It adopted the GPD/MCE training algorithm to train the HMM models for 100 final-dependent syllable initials and 40 syllable finals. The SBR method was used to compensate the speaker and channel effects. Besides, an RNN-based pre-classification scheme was employed to speed up the recognition search. A syllable recognition rate of 53.7% was achieved. This method was then used to implement an isolated-word recognizer for the prototype system to discriminate 1922 names of bank and insurance companies. Word recognition rates of 94.8% for top-1 and 97.9% for top-3 were achieved.