Refine
Document Type
- Part of a Book (25)
- Article (5)
- Conference Proceeding (5)
- Book (2)
Keywords
Publicationstate
- Zweitveröffentlichung (37) (remove)
Reviewstate
- (Verlags)-Lektorat (24)
- Peer-Review (11)
- (Verlags-)lektorat (1)
Publisher
- de Gruyter (17)
- Narr (5)
- European Language Resources Association (3)
- Wilhelm Fink (3)
- CLARIN (1)
- De Gruyter (1)
- Edinburgh University Press (1)
- Erich Schmidt (1)
- Klostermann (1)
- Sage Publishing (1)
Einleitung
(2023)
Einleitung
(2023)
Der Beitrag illustriert die Nutzung des Forschungs- und Lehrkorpus Gesprochenes Deutsch (FOLK) für interaktionslinguistische Fragestellungen anhand einer exemplarischen Studie. Zunächst werden die Stratifikation (Datenkomposition) des Korpus, das zugrundeliegende Datenmodell und dessen Annotationsebenen sowie Typen von Untersuchungsinteressen vorgestellt, für die das Korpus nutzbar ist. Im Hauptteil wird Schritt für Schritt anhand einer Studie zur Verwendung des Formats was heißt X in der sozialen Interaktion gezeigt, wie mit FOLK relevante Daten gefunden und analysiert werden können. Abschließend weisen wir auf einige Vorsichtsmaßnahmen bei der Benutzung des Korpus hin.
KonsortSWD ist das NFDI Konsortium für die Sozial-, Verhaltens-, Bildungs- und Wirtschaftswissenschaften. Für die äußerst vielfältigen Datentypen und Forschungsmethoden bauen die Beteiligten im Rahmen der NFDI eine bereits bestehende Forschungsdateninfrastruktur aus und ergänzen neue integrierende Dienste. Basis sind die heute 41 vom Rat für Sozial- und Wirtschaftsdaten akkreditierten Forschungsdatenzentren (FDZ). FDZ sind Spezialsammlungen zu jeweils spezifischen Forschungsdaten, z.B. aus der qualitativen Sozialforschung, und können so Forschende auf Basis einer ausführlichen Expertise zu diesen Daten beraten. Neben der Unterstützung der FDZ baut KonsortSWD auch neue Dienste in den Bereichen Datenproduktion, Datenzugang und Technische Lösungen auf.
The newest generation of speech technology caused a huge increase of audio-visual data nowadays being enhanced with orthographic transcripts such as in automatic subtitling in online platforms. Research data centers and archives contain a range of new and historical data, which are currently only partially transcribed and therefore only partially accessible for systematic querying. Automatic Speech Recognition (ASR) is one option of making that data accessible. This paper tests the usability of a state-of-the-art ASR-System on a historical (from the 1960s), but regionally balanced corpus of spoken German, and a relatively new corpus (from 2012) recorded in a narrow area. We observed a regional bias of the ASR-System with higher recognition scores for the north of Germany vs. lower scores for the south. A detailed analysis of the narrow region data revealed – despite relatively high ASR-confidence – some specific word errors due to a lack of regional adaptation. These findings need to be considered in decisions on further data processing and the curation of corpora, e.g. correcting transcripts or transcribing from scratch. Such geography-dependent analyses can also have the potential for ASR-development to make targeted data selection for training/adaptation and to increase the sensitivity towards varieties of pluricentric languages.
The newest generation of speech technology caused a huge increase of audio-visual data nowadays being enhanced with orthographic transcripts such as in automatic subtitling in online platforms. Research data centers and archives contain a range of new and historical data, which are currently only partially transcribed and therefore only partially accessible for systematic querying. Automatic Speech Recognition (ASR) is one option of making that data accessible. This paper tests the usability of a state-of-the-art ASR-System on a historical (from the 1960s), but regionally balanced corpus of spoken German, and a relatively new corpus (from 2012) recorded in a narrow area. We observed a regional bias of the ASR-System with higher recognition scores for the north of Germany vs. lower scores for the south. A detailed analysis of the narrow region data revealed – despite relatively high ASR-confidence – some specific word errors due to a lack of regional adaptation. These findings need to be considered in decisions on further data processing and the curation of corpora, e.g. correcting transcripts or transcribing from scratch. Such geography-dependent analyses can also have the potential for ASR-development to make targeted data selection for training/adaptation and to increase the sensitivity towards varieties of pluricentric languages.
This paper presents experiments on sentence boundary detection in transcripts of spoken dialogues. Segmenting spoken language into sentence-like units is a challenging task, due to disfluencies, ungrammatical or fragmented structures and the lack of punctuation. In addition, one of the main bottlenecks for many NLP applications for spoken language is the small size of the training data, as the transcription and annotation of spoken language is by far more time-consuming and labour-intensive than processing written language. We therefore investigate the benefits of data expansion and transfer learning and test different ML architectures for this task. Our results show that data expansion is not straightforward and even data from the same domain does not always improve results. They also highlight the importance of modelling, i.e. of finding the best architecture and data representation for the task at hand. For the detection of boundaries in spoken language transcripts, we achieve a substantial improvement when framing the boundary detection problem as a sentence pair classification task, as compared to a sequence tagging approach.