Systems, methods, and computer program products for machine translation are provided. In some implementations a system is provided. The system includes a language model including a collection of n-grams from a corpus, each n-gram having a corresponding relative frequency in the corpus and an order n corresponding to a number of tokens in the n-gram, each n-gram corresponding to a backoff n-gram having an order of n-1 and a collection of backoff scores, each backoff score associated with an n-gram, the backoff score determined as a function of a backoff factor and a relative frequency of a corresponding backoff n-gram in the corpus.

/pdf/large-language-models-in-machine-translation-3wqubjuzrp.pdf

Large Language Models in Machine Translation

Automatic speech recognition for under-resourced languages: A survey

A multi-modal human computer interface (HCI) receives a plurality of available information inputs concurrently, or serially, and employs a subset of the inputs to determine or infer user intent with respect to a communication or information goal. Received inputs are respectively parsed, and the parsed inputs are analyzed and optionally synthesized with respect to one or more of each other. In the event sufficient information is not available to determine user intent or goal, feedback can be provided to the user in order to facilitate clarifying, confirming, or augmenting the information inputs.

/pdf/speech-centric-multimodal-user-interface-design-in-mobile-jiqw1wzszt.pdf

Speech-centric multimodal user interface design in mobile technology

Written for students of linguistics, applied linguistics and speech therapy, this dictionary covers over 2,000 terms in phonetics and phonology. In addition to providing a comprehensive, yet concise, guide to an enormous number of individual terms, it also includes an explanation of the most important theoretical approaches to phonology. Its usefulness as a reference tool is further enhanced by the inclusion of pronunciations, notational devices and symbols, earliest sources of terms, suggestions for further reading, and advice with regard to usage. The wide range of topics explained include: * Classical phonology, including American Structuralism and the Prague School * Contemporary approaches, including Autosegmental Phonology, Metrical Phonology, Dependency Phonology, Government Phonology and Lexical Phonology * Prosodic ideas in phonology, both traditional and contemporary ^ * * historical phonology * Intonation and tonology This dictionary devotes space to the various theoretical approaches in proportion to their importance, but it concentrates most heavily on non-theory-bound descriptive terminology. It will remain a definitive reference for years to come.

语音学和音系学词典 = A dictionary of phonetics and phonology

Systems, methods and apparatus for collecting data and presenting media to a user are provided. The systems generally includes a data gathering module associated with an electronic device. The data gathering module communicates gathered data to a management module, which manages at least one user profile based on the gathered data. The management module may select media for presentation to a user based on the user profile, and the selected media may be displayed to the user via a media output device co-located with the user, such as a display of the user's mobile electronic device or a television, computer, billboard or other display co-located with the user. Related methods are also provided.

Methods, systems and apparatus for delivery of media

An apparatus generates a statistical class sequence model called A class bi-multigram model from input training strings of discrete-valued units, where bigram dependencies are assumed between adjacent variable length sequences of maximum length N units, and where class labels are assigned to the sequences. The number of times all sequences of units occur are counted, as well as the number of times all pairs of sequences of units co-occur in the input training strings. An initial bigram probability distribution of all the pairs of sequences is computed as the number of times the two sequences co-occur, divided by the number of times the first sequence occurs in the input training string. Then, the input sequences are classified into a pre-specified desired number of classes. Further, an estimate of the bigram probability distribution of the sequences is calculated by using an EM algorithm to maximize the likelihood of the input training string computed with the input probability distributions. The above processes are then iteratively performed to generate statistical class sequence model.

Apparatus for generating a statistical sequence model called class bi-multigram model with bigram dependencies assumed between adjacent sequences

A command for application program is generated based on both a movement of a cursor on a display unit depending upon operation of a pointing device and a voice produced in parallel to the operation when the pointing device is operated to select an object being displayed on the display unit connected to a computer. Particularly, if a moving speed of the cursor in a region of the object has a local minimum value which is less than a predetermined speed, the object is selected as a referent candidate for the voice containing a demonstrative word. If a plurality of referent candidates each having the local minimum value less than the predetermined speed are present for the voice containing the demonstrative word, such object is recognized as a referent for the voice that a time period during when the cursor moves in a region overlaps at maximum with a time period during when the voice is produced.

Multimodal information inputting method and apparatus for embodying the same

We present a portable translator that recognizes and translates phrases on signboards and menus as captured by a built-in camera. This system can be used on PDAs or mobile phones and resolves the difficulty of inputting some character sets such as Japanese and Chinese if the user doesn't know their readings. Through the high speed mobile network, small images of signboards can be quickly sent to the recognition and translation server. Since the server runs state of the art recognition and translation technology and huge dictionaries, the proposed system offers more accurate character recognition and machine translation.

Portable Translator Capable of Recognizing Characters on Signboard and Menu Captured by its Built-in Camera

This paper describes an approach to HMM-based expressive speech synthesis which does not require any supervised labeling process for emphasis context. We use appealing-style speech whose sentences were taken from real domains. To reduce the cost for labeling speech data with an emphasis context for the model training, we propose an unsupervised labeling technique of the emphasis context based on the difference between original and generated F0 patterns of training sentences. Although the criterion for the emphasis labeling is quite simple, subjective evaluation results reveal that the unsupervised labeling is comparable to the labeling conducted carefully by a human in terms of speech naturalness and emphasis reproducibility. Index Terms: HMM-based speech synthesis, expressive speech, emphasis expression, unsupervised labeling, F0 generation

/pdf/hmm-based-emphatic-speech-synthesis-using-unsupervised-1ti70tyv7n.pdf

HMM-based emphatic speech synthesis using unsupervised context labeling

Statistical language modeling requires a large corpus for the application domain. When a large corpus is not available, the language model adaptation technique has often been used in the speech recognition research domain. This adaptation needs only a small corpus of the application domain (the "target corpus") and the corpus should be written in the language of the model. However, it is sometimes difficult to collect even a small corpus, especially of spoken language, due to its high cost. To address this problem, this paper proposes a novel scheme that generates a small target corpus in the language of the model by machine translation of the target corpus in another language. As information about adjacent words, which is necessary for a statistical language model, is stored in the translation knowledge, it can be extracted by machine translation and used for adaptation. Experiments showed that the language model improvement was about half of that which was obtained with a human collected corpus, and this provided some initial proof of the concept experiments.

Hideharu Nakajima

Papers

Apparatus for generating a statistical sequence model called class bi-multigram model with bigram dependencies assumed between adjacent sequences

Multimodal information inputting method and apparatus for embodying the same

Portable Translator Capable of Recognizing Characters on Signboard and Menu Captured by its Built-in Camera

HMM-based emphatic speech synthesis using unsupervised context labeling

Language model adaptation with additional text generated by machine translation