Grammar Recognizers

Notes:

A grammar recognizer is a software tool that is used to analyze and parse spoken or written language according to the rules of a specific grammar. Grammar recognizers are used in a variety of applications, including speech recognition systems, natural language processing (NLP) systems, and chatbots or virtual assistants.

In speech recognition systems, grammar recognizers are used to analyze and understand spoken language and convert it into text. This can involve identifying individual words and phrases, as well as determining the grammatical structure of the sentence and the relationships between words and phrases.

In NLP systems, grammar recognizers are used to analyze and understand written language and extract meaning from it. This can involve identifying the parts of speech in a sentence, determining the grammatical structure of the sentence, and identifying relationships between words and phrases.

In chatbots and virtual assistants, grammar recognizers are used to analyze and understand the input from users and determine how to respond. This can involve identifying the intended meaning of the user’s input, determining the appropriate response based on the rules of the chatbot’s grammar, and generating a response in natural language.

Grammar recognizers are related to the Speech Recognition Grammar Specification (SRGS) in that they are used to interpret and recognize spoken language input based on the rules and structures defined in an SRGS document.

SRGS is a W3C recommendation that defines a markup language for representing the grammar of a spoken language. It specifies a way to describe the words, phrases, and rules that make up a grammar, as well as the structure and organization of that grammar.

Grammar recognizers are software programs that use SRGS documents to interpret and recognize spoken language input. They do this by comparing the spoken input to the rules and structures defined in the SRGS document and determining if the input matches the expected grammar. If the input matches the grammar, the grammar recognizer will output the recognized text. If the input does not match the grammar, the grammar recognizer will typically output an error or a message indicating that the input was not recognized.

Grammar recognizers are often used in speech recognition systems to improve the accuracy and reliability of the recognition process. By specifying the expected grammar in an SRGS document, the system can more effectively filter out input that does not conform to the expected grammar, resulting in more accurate and reliable recognition results.

Wikipedia:

See also:

Formal Grammar

  1. Capelletti, M. (n.d.). Computer Science and Engineering Oregon Graduate Institute PO Box 91000 Portland OR 97291.
  2. Carmichael, J. N. (2001). The Implementation of a Speech Interface for The THISL Information Retrieval Engine. [PDF] from shef.ac.uk. Retrieved from http://dcs.shef.ac.uk/~james/reports/james_carmichael_msc_project_report.pdf
  3. Fernandez, F., Sama, V., D’Haro, L. F., & others. (2004). Implementation of dialog applications in an open-source VoiceXML platform. In Proceedings of the 8th International Conference on Spoken Language Processing (ICSLP 2004) (Vol. 2, pp. 1349-1352).
  4. Leung, K. Y., Mak, M. W., Siu, M., & Young, S. (2005). Speaker Verification via Articulatory Feature-based Conditional Pronunciation Modeling with Vowel and Consonant Mixture Models. In Proceedings of the 9th European Conference on … Retrieved from https://www.isca-speech.org/archive/archive_papers/eurospeech_2005/e05_0499.pdf
  5. Moore, R., Pereira, F., & others. (1989). Integrating speech and natural-language processing. Speech and Natural Language Processing, 1989. ACL-89, 79-85.
  6. Paul, D. B. (1992). The Lincoln large-vocabulary HMM CSR. Proceedings of the workshop on Speech and Natural … Retrieved from https://www.isca-speech.org/archive_open/archive_papers/snp93_201.pdf
  7. Pawate, B. I., & Doddington, G. R. (1989). Implementation of a hidden Markov model-based layered grammar recognizer. Acoustics, Speech, and Signal Processing, 1989. ICASSP-89, Glasgow, Scotland, 4, 1377-1380.
  8. Picone, J. W. (1989). A phonetic vocoder. Acoustics, Speech, and Signal Processing, 1989. ICASSP-89, Glasgow, Scotland, 4, 2423-2426.
  9. Shanmugham, S. S. (2006). Media Resource Control Protocol Version 2 (MRCPv2). [PDF] from 194.146.105.14. Retrieved from http://194.146.105.14/html/draft-ietf-speechsc-mrcpv2-11
  10. Spiertz, P. P. (2003). [PS] Towards a Grammar Workbench for AGFL [PS] from ru.nl. Retrieved from https://www.cs.ru.nl/~peter/sgngwb98/papers/11-abstract.pdf
  11. Surace, K. J., White, G. M., Reeves, B. B., Nass, C. I., & Lieberman, H. (2000). Voice user interface with personality. US Patent 6,081,938. Retrieved from https://patents.google.com/patent/US6081938A/en
  12. Thomas, J. C. (1982). An Advanced Human Interface for Computer Assisted Instruction in Propulsion Engineering. In Gathering Information for Problem Formulation. Retrieved from https://smartech.gatech.edu/bitstream/handle/1853/2074/A-216.pdf
  13. Walker, W. D., Hunt, A. J., Adams, S. J., & others. (2002). System and method for referencing object instances and invoking methods on those object instances from within a speech recognition grammar. US Patent 6,434,529.
  14. Zhang, S. (2008). Articulatory-feature based pronunciation modelling for high-level speaker verification. Retrieved from https://repository.lib.polyu.edu.hk/jspui/handle/123456789/5911
  15. Zhang, S. X., Mak, M. W., Siu, M., & others. (2007). Speaker verification via high-level feature based phonetic-class pronunciation modeling. IEEE Transactions on Audio, Speech, and Language Processing, 15(1), 86-95.
  16. Zhang, S. X., Mak, M. W., & Siu, M. (2008). Articulatory-feature based sequence kernel for high-level speaker verification. In Proceedings of the International Conference on Machine Learning and Cybernetics (ICMLC 2008) (Vol. 2, pp. 660-664).