Search

Automatic speech recognition (ASR) has become a valuable tool in large document production environments like medical dictation. While manual post-processing is still needed for correcting speech recognition errors and for creating documents which adhere to various stylistic and formatting conventions, a large part of the document production process is carried out by the ASR system. For improving the quality of the system output, knowledge about the multi-layered relationship between the dictated texts and the final documents is required. Thus, typical speech-recognition errors can be avoided, and proper style and formatting can be anticipated in the ASR part of the document production process. Yet - while vast amounts of recognition results and manually edited final reports are constantly being produced - the error-free literal transcripts of the actually dictated texts are a scarce and costly resource because they have to be created by manually transcribing the audio files. To obtain large corpora of literal transcripts for medical dictation, we propose a method for automatically reconstructing them from draft speech-recognition transcripts plus the corresponding final medical reports. The main innovative aspect of our method is the combination of two independent knowledge sources: phonetic information for the identification of speech-recognition errors and semantic information for detecting post-editing concerning format and style. Speech recognition results and final reports are first aligned, then properly matched based on semantic and phonetic similarity, and finally categorised and selectively combined into a reconstruction hypothesis. This method can be used for various applications in language technology, e.g., adaptation for ASR, document production, or generally for the development of parallel text corpora of non-literal text resources. In an experimental evaluation, which also includes an assessment of the quality of the reconstructed transcripts compared to manual transcriptions, the described method results in a relative word error rate reduction of 7.74% after retraining the standard language model with reconstructed transcripts.

Automatic processing of medical dictations poses a significant challenge. We approach the problem by introducing a statistical framework capable of identifying types and boundaries of sections, lists and other structures occurring in a dictation, thereby gaining explicit knowledge about the function of such elements. Training data is created semi-automatically by aligning a parallel corpus of corrected medical reports and corresponding transcripts generated via automatic speech recognition. We highlight the properties of our statistical framework, which is based on conditional random fields (CRFs) and implemented as an efficient, publicly available toolkit. Finally, we show that our approach is effective both under ideal conditions and for real-life dictation involving speech recognition errors and speech-related phenomena such as hesitation and repetitions.

This paper describes the language component of FASTY, a text prediction system designed to improve text input efficiency for disabled users. The FASTY language component is based on state-of-the-art n-gram-based word-level and part-of-speech-level prediction and on a number of innovative modules (morphological analysis, collocation-based prediction, compound prediction) that are meant to enhance performance in languages other than English. Together with its modular architecture, these novel techniques make it adaptable to a wide range of languages without sacrificing performance. Currently, versions for Dutch, German, French, Italian, and Swedish are supported. The system can be parameterized to be used with different user interfaces and for a range of different applications. In this paper, we discuss each of the modules in detail and we present a series of experimental evaluations of the system.

In this paper, we report about some preliminary experiments in which we tried to improve the performance of a state-of-the-art Predictive Typing system for the German language by adding a collocation-based prediction component. This component tries to exploit the fact that texts have a topic and are semantically coherent. Thus, the appearance in a text of a certain word can be a cue that other, semantically related words are likely to appear soon. The collocation-based module exploits this kind of topical/semantic relatedness by relying on statistics about the co-occurrence of words within a large window of text in the training corpus. Our current experimental results indicate that using the collocation-based prediction module has a small but consistent positive effect on the performance of the system.

We present an algorithm that takes an unannotated corpus as its input, and returns a ranked list of probable morphologically related pairs as its output. The algorithm tries to discover morphologically related pairs by looking for pairs that are both orthographically and semantically similar, where orthographic similarity is measured in terms of minimum edit distance, and semantic similarity is measured in terms of mutual information. The procedure does not rely on a morpheme concatenation model, nor on distributional properties of word substrings (such as affix frequency). Experiments with German and English input give encouraging results, both in terms of precision (proportion of good pairs found at various cutoff points of the ranked list), and in terms of a qualitative analysis of the types of morphological patterns discovered by the algorithm., Comment: 10 pages, to appear in the Proceedings of the Morphology/Phonology Learning Workshop of ACL-02

In word prediction systems for augmentative and alternative communication (AAC), productive word-formation processes such as compounding pose a serious problem. We present a model that predicts German nominal compounds by splitting them into their modifier and head components, instead of trying to predict them as a whole. The model is improved further by the use of class-based modifier-head bigrams constructed using semantic classes automatically extracted from a corpus. The evaluation shows that the split compound model with class bigrams leads to an improvement in keystroke savings of more than 15% over a no split compound baseline model. We also present preliminary results obtained with a word prediction model integrating compound and simple word prediction.

Communication and information exchange is a vital factor in human society. Communication disorders severely influence the quality of life. Whereas experienced typists will produce some 300 keystrokes per minute, persons with motor impairments achieve only much lower rates. Predictive typing systems for English speaking areas have proven useful andefficien t, but for all other European languages there exist no predictive typing programs powerful enough to substantially improve the communication rate andthe IT access for disabledp ersons. FASTY aims at offering a communication support system significantly increasing typing speed, adaptable to users with different language and strongly varying needs. In this way the large group of non-English-speaking disabled citizens will be supported in living a more independent and self determined life.

We describe the development of a generator for German built by reusing and adapting existing linguistic data and software. Reusability is crucial for the successful application of NLP techniques to real-life problems since it helps to cut down on both development and adaptation effort. However, combining resources not designed to work together is not trivial. We describe the problems arising when integrating three preexisting resources (FUF, a unification-based generator, an HPSG Grammar for German, and X2MorF, a two-level morphology component) and the adaptations necessary to come up with a wide coverage tactical generator for German.

We present an integrated architecture for word-level and sentence-level processing in a unification-based paradigm. The core of the system is a CLP implementation of a unification engine for feature structures supporting relational values. In this framework an HPSG-style grammar is implemented. Word-level processing uses X2MorF, a morphological component based on an extended version of two-level morphology. This component is tightly integrated with the grammar as a relation. The advantage of this approach is that morphology and syntax are kept logically autonomous while at the same time minimizing interface problems., 7 pages, LaTeX, uses fullname.sty, to appear in COLING'95

Datenbank-DIALOG is a German language interface for relational databases. Its goal is to provide easy and efficient access to the contents of such a database for casual users. The user has only to type in his or her question in plain German and the system will respond properly. This is much easier than using a formal query language and more flexible than a menu-type or graphic interaction. Datenbank-DIALOG was designed to be as portable as possible, in particular with regard to the application domains as well as the database management system and the operating system of the supporting computer. Equally important was good coverage of the subset of German relevant for database retrieval. German was chosen because we feel that at present it is not feasible to design a parser adaptable to different natural languages, at least not in a satisfying manner. As a result of this specialization Datenbank-DIALOG imposes almost no syntactic restrictions on the user, and care has been taken to make the semanti...

One main task of a natural language interface to a relational database is to map the semantic structures created by the parser onto database relations by translating them into statements of the query language. In this paper we present a method which employs an intermediate logical representation based on the theory of Generalized Quantifiers. These theoretical foundations provide the framework for an effective and straightforward implementation.