Summary of the paper
| Title | Example-Based Automatic Phonetic Transcription |
|---|---|
| Authors | Christina Leitner, Martin Schickbichler and Stefan Petrik |
| Abstract | Current state-of-the-art systems for automatic phonetic transcription (APT) aremostly phone recognizers based on Hidden Markov models (HMMs). We present adifferent approach for APT especially designed for transcription with a largeinventory of phonetic symbols. In contrast to most systems which aremodel-based, our approach is non-parametric using techniques derived fromconcatenative speech synthesis and template-based speech recognition. Thisexample-based approach not only produces draft transcriptions that just need tobe corrected instead of created from scratch but also provides a validationmechanism for ensuring consistency within the corpus. Implementations of thistranscription framework are available as standalone Java software and extensionto the ELAN linguistic annotation software. The transcription system was testedwith audio files and reference transcriptions from the Austrian PronunciationDatabase (ADABA) and compared to an HMM-based system trained on the same dataset. The example-based and the HMM-based system achieve comparable phonerecognition rates. A combination of rule-based and example-based APT in aconstrained phone recognition scenario returned the best results. |
| Language | Corpus (creation, annotation, etc.) |
| Topics | Tools, systems, applications, Phonetic Databases, Phonology, Corpus (creation, annotation, etc.) |
Full paper  |
Example-Based Automatic Phonetic Transcription |
| Bibtex | @InProceedings{LEITNER10.299, author = {Christina Leitner, Martin Schickbichler and Stefan Petrik}, title = {Example-Based Automatic Phonetic Transcription}, booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)}, year = {2010}, month = {may}, date = {19-21}, address = {Valletta, Malta}, editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias}, publisher = {European Language Resources Association (ELRA)}, isbn = {2-9517408-6-7}, language = {english} } |