Parametric trajectory models explicitly represent the temporal evolution of the speech features as a Gaussian process with time-varying parameters. HMMs are a special case of such models, one in which the trajectory constraints in the speech segment are ignored by the assumption of conditional independence across frames within the segment. In this paper, we investigate in detail some extensions to our trajectory modeling approach aimed at improving LVCSR performance: (i) improved modeling of mixtures of trajectories via better initialization, (ii) modeling of context dependence, and (iii) improved segment boundaries by means of search. We will present results in terms of both phone classi cation and recognition accuracy on the Switchboard corpus.
{"title":"Parametric trajectory mixtures for LVCSR","authors":"M. Siu, R. Iyer, H. Gish, Carl Quillen","doi":"10.21437/ICSLP.1998-685","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-685","url":null,"abstract":"Parametric trajectory models explicitly represent the temporal evolution of the speech features as a Gaussian process with time-varying parameters. HMMs are a special case of such models, one in which the trajectory constraints in the speech segment are ignored by the assumption of conditional independence across frames within the segment. In this paper, we investigate in detail some extensions to our trajectory modeling approach aimed at improving LVCSR performance: (i) improved modeling of mixtures of trajectories via better initialization, (ii) modeling of context dependence, and (iii) improved segment boundaries by means of search. We will present results in terms of both phone classi cation and recognition accuracy on the Switchboard corpus.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130186963","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Cheesman, K. Smilsky, T. Major, F. Lewis, L. Boorman
A sample of 209 adults ranging from 20 to 79 years of age were studied to measure speech communication profiles as a function of age in persons who did not identify themselves as hearing impaired. The study was conducted in order to evaluate age-related speech percepton abilities and ccammmication profiles in a population who do not present for hearing assessment and who are not included in census statistics as having hearing problems. Audiometric assessment, demographic and hearing history self-reports, speech reception thresholds, consonant discrimination perception in quiet and noise, and the Ccumnunication Profile for the Hearing Impaired (CPHI) were the in.ements used to develop speech communication profiles. Hearing performance decreased with increased age. However, despite self-reports of no hearing impairment, many subjects over age 50 had audiometric thresholds that indicated hearing impairment. The responses to the CPHI were correlated to audiometric thresholds, but also to the age of the respondent, when hearing thresholds had been controlled statistically. A comparison of CPHI responses f?om this study and that of two other samples in clinical populations revealed only slightly different patterns of behaviour in the present sample when co&o&d with communication difficulties.
{"title":"Speech communication profiles across the adult lifespan: persons without self-identified hearing impairment","authors":"M. Cheesman, K. Smilsky, T. Major, F. Lewis, L. Boorman","doi":"10.21437/ICSLP.1998-790","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-790","url":null,"abstract":"A sample of 209 adults ranging from 20 to 79 years of age were studied to measure speech communication profiles as a function of age in persons who did not identify themselves as hearing impaired. The study was conducted in order to evaluate age-related speech percepton abilities and ccammmication profiles in a population who do not present for hearing assessment and who are not included in census statistics as having hearing problems. Audiometric assessment, demographic and hearing history self-reports, speech reception thresholds, consonant discrimination perception in quiet and noise, and the Ccumnunication Profile for the Hearing Impaired (CPHI) were the in.ements used to develop speech communication profiles. Hearing performance decreased with increased age. However, despite self-reports of no hearing impairment, many subjects over age 50 had audiometric thresholds that indicated hearing impairment. The responses to the CPHI were correlated to audiometric thresholds, but also to the age of the respondent, when hearing thresholds had been controlled statistically. A comparison of CPHI responses f?om this study and that of two other samples in clinical populations revealed only slightly different patterns of behaviour in the present sample when co&o&d with communication difficulties.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"161 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133893507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we propose a novel implementation of a minimax decision rule for continuous density hidden Markov model based robust speech recognition. By combining the idea of the minimax decision rule with a normal Viterbi search, we derive a recursive minimax search algorithm, where the minimax decision rule is repetitively applied to determine the partial paths during the search procedure. Because of its intrinsic nature of a recursive search, the proposed method can be easily extended to perform contin-uos speech recognition. Experimental results on Japanese isolated digits and TIDIGITS, where the mismatch between training and testing conditions is caused by additive white Gaussian noise, show the viability and e(cid:14)ciency of the proposed minimax search algorithm.
{"title":"A minimax search algorithm for CDHMM based robust continuous speech recognition","authors":"Hui Jiang, K. Hirose, Qiang Huo","doi":"10.21437/ICSLP.1998-304","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-304","url":null,"abstract":"In this paper, we propose a novel implementation of a minimax decision rule for continuous density hidden Markov model based robust speech recognition. By combining the idea of the minimax decision rule with a normal Viterbi search, we derive a recursive minimax search algorithm, where the minimax decision rule is repetitively applied to determine the partial paths during the search procedure. Because of its intrinsic nature of a recursive search, the proposed method can be easily extended to perform contin-uos speech recognition. Experimental results on Japanese isolated digits and TIDIGITS, where the mismatch between training and testing conditions is caused by additive white Gaussian noise, show the viability and e(cid:14)ciency of the proposed minimax search algorithm.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134135137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper describes some of the implementation details of the Nozomi" 1 stack decoder for LVCSR. The decoder was tested on a Japanese Newspaper Dictation Task using a 5000 word vocabulary. Using continuous density acoustic models with 2000 and 3000 states trained on the JNAS/ASJ corpora and a 3-gram LM trained on the RWC text corpus, both models provided by the IPA group [7], it was possible to reach more than 95% word accuracy on the standard test set. With computationally cheap acoustic models we could achieve around 89% accuracy in nearly realtime on a 300 Mhz Pentium II. Using a disk-based LM the memory usage could be optimized to 4 MB in total.
{"title":"Nozomi - a fast, memory-efficient stack decoder for LVCSR","authors":"M. Schuster","doi":"10.21437/ICSLP.1998-627","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-627","url":null,"abstract":"This paper describes some of the implementation details of the Nozomi\" 1 stack decoder for LVCSR. The decoder was tested on a Japanese Newspaper Dictation Task using a 5000 word vocabulary. Using continuous density acoustic models with 2000 and 3000 states trained on the JNAS/ASJ corpora and a 3-gram LM trained on the RWC text corpus, both models provided by the IPA group [7], it was possible to reach more than 95% word accuracy on the standard test set. With computationally cheap acoustic models we could achieve around 89% accuracy in nearly realtime on a 300 Mhz Pentium II. Using a disk-based LM the memory usage could be optimized to 4 MB in total.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133904476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Contemporary psycholinguistic models suggest that there may be dual routes operating in phonetic encoding: a direct route which uses stored syllabic units, and an indirect route which relies on the on-line assembly of sub-syllabic units. The more computationally efficient direct route is more likely to be used for high frequency words, while the indirect route is most likely to be used for novel or low frequency words. We suggest that the acquired neurological disorder of apraxia of speech (AOS), provides a window to speech encoding mechanisms and that the disorder represents an impairment of direct route encoding mechanisms and, therefore, a reliance on indirect mechanisms. We report an investigation of the production of high and low frequency words across three subject groups: non-brain damaged control (NBDC, N=3); brain damaged control (BDC, N=3) and speakers with AOS (N=4). The results are presented and discussed within the dual-route phonetic encoding hypothesis.
{"title":"Evidence of dual-route phonetic encoding from apraxia of speech: implications for phonetic encoding models","authors":"R. Varley, S. Whiteside","doi":"10.21437/ICSLP.1998-789","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-789","url":null,"abstract":"Contemporary psycholinguistic models suggest that there may be dual routes operating in phonetic encoding: a direct route which uses stored syllabic units, and an indirect route which relies on the on-line assembly of sub-syllabic units. The more computationally efficient direct route is more likely to be used for high frequency words, while the indirect route is most likely to be used for novel or low frequency words. We suggest that the acquired neurological disorder of apraxia of speech (AOS), provides a window to speech encoding mechanisms and that the disorder represents an impairment of direct route encoding mechanisms and, therefore, a reliance on indirect mechanisms. We report an investigation of the production of high and low frequency words across three subject groups: non-brain damaged control (NBDC, N=3); brain damaged control (BDC, N=3) and speakers with AOS (N=4). The results are presented and discussed within the dual-route phonetic encoding hypothesis.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133996680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a context-based analysis of the intervals between different speakers’ utterances in a corpus of taskoriented dialogue (the Human Communication Research Centre’s Map Task Corpus. See Anderson et al. 1991). In the analysis, we assessed the relationship between inter-speaker intervals and various contextual factors, such as the effects of eye contact, the presence of conversational game boundaries, the category of move in an utterance, and the degree of experience with the task in hand. The results of the analysis indicated that the main factors which gave rise to significant differences in inter-speaker intervals were those which related to decision-making and planning the greater the amount of planning, the greater the inter-speaker interval. Differences between speakers were also found to be significant, although this effect did not necessarily interact with all other effects. These results provide unique and useful data for the improved effectiveness of dialogue systems.
本文对任务型对话语料库(人类交际研究中心地图任务语料库)中不同说话人的话语间隔进行了基于语境的分析。参见Anderson et al. 1991)。在分析中，我们评估了说话人之间的间隔与各种上下文因素之间的关系，例如眼神接触的影响、会话游戏边界的存在、话语中的移动类别以及手头任务的经验程度。分析结果表明，引起语间间隔显著差异的主要因素是与决策和计划相关的因素，计划量越大，语间间隔越大。说话者之间的差异也被发现是显著的，尽管这种影响不一定与所有其他影响相互作用。这些结果为提高对话系统的有效性提供了独特而有用的数据。
{"title":"An analysis of the timing of turn-taking in a corpus of goal-oriented dialogue","authors":"Matthew Bull, M. Aylett","doi":"10.21437/ICSLP.1998-81","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-81","url":null,"abstract":"This paper presents a context-based analysis of the intervals between different speakers’ utterances in a corpus of taskoriented dialogue (the Human Communication Research Centre’s Map Task Corpus. See Anderson et al. 1991). In the analysis, we assessed the relationship between inter-speaker intervals and various contextual factors, such as the effects of eye contact, the presence of conversational game boundaries, the category of move in an utterance, and the degree of experience with the task in hand. The results of the analysis indicated that the main factors which gave rise to significant differences in inter-speaker intervals were those which related to decision-making and planning the greater the amount of planning, the greater the inter-speaker interval. Differences between speakers were also found to be significant, although this effect did not necessarily interact with all other effects. These results provide unique and useful data for the improved effectiveness of dialogue systems.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131553624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, the problem of robust speech recognition has been considered. Our approach is based on the noise reduction of the parameters that we use for recognition, that is, the Mel-based cepstral coefficients. A Temporal-Correlation-Based Recurrent Multilayer Neural Network (TCRMNN) for noise reduction in the cepstral domain is used in order to get less-variant parameters to be useful for robust recognition in noisy environments. Experiments show that the use of the enhanced parameters using such an approach increases the recognition rate of the continuous speech recognition (CSR) process. The HTK Hidden Markov Model Toolkit was used throughout. Experiments were done on a noisy version of the TIMIT database. With such a pre-processing noise reduction technique in the front-end of the HTK-based continuous speech recognition system (CSR) system, improvements in the recognition accuracy of about 17.77% and 18.58% using single mixture monophones and triphones, respectively, have been obtained at a moderate SNR of 20 dB.
{"title":"Robust automatic speech recognition by the application of a temporal-correlation-based recurrent multilayer neural network to the mel-based cepstral coefficients","authors":"M. Héon, H. Tolba, D. O'Shaughnessy","doi":"10.21437/ICSLP.1998-328","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-328","url":null,"abstract":"In this paper, the problem of robust speech recognition has been considered. Our approach is based on the noise reduction of the parameters that we use for recognition, that is, the Mel-based cepstral coefficients. A Temporal-Correlation-Based Recurrent Multilayer Neural Network (TCRMNN) for noise reduction in the cepstral domain is used in order to get less-variant parameters to be useful for robust recognition in noisy environments. Experiments show that the use of the enhanced parameters using such an approach increases the recognition rate of the continuous speech recognition (CSR) process. The HTK Hidden Markov Model Toolkit was used throughout. Experiments were done on a noisy version of the TIMIT database. With such a pre-processing noise reduction technique in the front-end of the HTK-based continuous speech recognition system (CSR) system, improvements in the recognition accuracy of about 17.77% and 18.58% using single mixture monophones and triphones, respectively, have been obtained at a moderate SNR of 20 dB.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131757815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Phonetic vocoding is one of the methods for coding speech below 1000 bit/s. The transmitter stage includes a phone recogniser whose index is transmitted together with prosodic information such as duration, energy and pitch variation. This type of coder does not transmit spectral speaker characteristics and speaker recognisability thus becomes a major problem. In our previous work, we adapted a speaker modification strategy to minimise this problem, modifying a codebook to match the spectral characteristics of the input speaker. This is done at the cost of transmitting the LSP averages computed for vowel and glide phones. This paper presents new codebook generation strategies, with gender dependence and interpolation frames, that lead to better speaker recognisability and speech quality. Relatively to our previous work, some effort was also devoted to deriving more efficient quantization methods for the speakerspecific information, that considerably reduced the average bit rate, without quality degradation. For the CD-ROM version, a set of audio files is also included.
{"title":"Improving speaker recognisability in phonetic vocoders","authors":"C. Ribeiro, I. Trancoso","doi":"10.21437/ICSLP.1998-396","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-396","url":null,"abstract":"Phonetic vocoding is one of the methods for coding speech below 1000 bit/s. The transmitter stage includes a phone recogniser whose index is transmitted together with prosodic information such as duration, energy and pitch variation. This type of coder does not transmit spectral speaker characteristics and speaker recognisability thus becomes a major problem. In our previous work, we adapted a speaker modification strategy to minimise this problem, modifying a codebook to match the spectral characteristics of the input speaker. This is done at the cost of transmitting the LSP averages computed for vowel and glide phones. This paper presents new codebook generation strategies, with gender dependence and interpolation frames, that lead to better speaker recognisability and speech quality. Relatively to our previous work, some effort was also devoted to deriving more efficient quantization methods for the speakerspecific information, that considerably reduced the average bit rate, without quality degradation. For the CD-ROM version, a set of audio files is also included.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131802616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The relative contributions of segmental versus prosodic factors to the perceived naturalness of synthetic speech was measured by transplanting prosody between natural speech and the output of a diphone synthesizer. A small corpus was created containing matched sentence pairs wherein one member of the pair was a natural utterance and the other was a synthetic utterance generated with diphone data from the same talker. Two additional sentences were formed from each sentence pair by transplanting the prosodic structure between the natural and synthetic members of each pair. In two listening experiments subjects were asked to (a) classify each sentence as “natural” or “synthetic, or (b) rate the naturalness of each sentence. Results showed that the prosodic information was more important than segmental information in both classification and ratings of naturalness.
{"title":"Prosodic vs. segmental contributions to naturalness in a diphone synthesizer","authors":"H. Bunnell, S. Hoskins, Debra Yarrington","doi":"10.21437/ICSLP.1998-15","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-15","url":null,"abstract":"The relative contributions of segmental versus prosodic factors to the perceived naturalness of synthetic speech was measured by transplanting prosody between natural speech and the output of a diphone synthesizer. A small corpus was created containing matched sentence pairs wherein one member of the pair was a natural utterance and the other was a synthetic utterance generated with diphone data from the same talker. Two additional sentences were formed from each sentence pair by transplanting the prosodic structure between the natural and synthetic members of each pair. In two listening experiments subjects were asked to (a) classify each sentence as “natural” or “synthetic, or (b) rate the naturalness of each sentence. Results showed that the prosodic information was more important than segmental information in both classification and ratings of naturalness.","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132637230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study investigated the perceptual properties of second language learners in acquiring second language phonemes. The case where the relation between two phonemes of a second language and those of a native language changes according to following vowels was studied. The perceptual properties of Russians with regards to Japanese fricatives were examined. In the perception test, the confusion of [ (cid:219) o] with [so] was very large. This phenomenon could be caused by the difference between the transition onset time from [s’] to vowels and that from the other consonants to vowels. It is considered that, in the case of following vowel [a] and [o], Russians equated Japanese [s] and [ (cid:219) ] with Russian [s] and [s'] respectively. However, in the case of [u], they did not equate them in such a manner. This is probably because the acoustic properties of Japanese [ (cid:149) ] are very different from those of Russian [u].
{"title":"Perceptual properties of Russians with Japanese fricatives","authors":"S. Funatsu, S. Kiritani","doi":"10.21437/ICSLP.1998-719","DOIUrl":"https://doi.org/10.21437/ICSLP.1998-719","url":null,"abstract":"This study investigated the perceptual properties of second language learners in acquiring second language phonemes. The case where the relation between two phonemes of a second language and those of a native language changes according to following vowels was studied. The perceptual properties of Russians with regards to Japanese fricatives were examined. In the perception test, the confusion of [ (cid:219) o] with [so] was very large. This phenomenon could be caused by the difference between the transition onset time from [s’] to vowels and that from the other consonants to vowels. It is considered that, in the case of following vowel [a] and [o], Russians equated Japanese [s] and [ (cid:219) ] with Russian [s] and [s'] respectively. However, in the case of [u], they did not equate them in such a manner. This is probably because the acoustic properties of Japanese [ (cid:149) ] are very different from those of Russian [u].","PeriodicalId":117113,"journal":{"name":"5th International Conference on Spoken Language Processing (ICSLP 1998)","volume":"17 3 Pt 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130780388","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：