WebMAUS-Basic: Automatic phonetic labelling & segmentation of a single German recording with text.

 

Interviews and conversation are often recorded and later transcribed. It is difficult in longer recordings to locate the precise section in which a phoneme, word or sentence is uttered, for example to find out about stress or hesitations uttered by the speaker. The web service WebMAUS Basic available in the CLARIN infrastructure alows to automatically combine audio recordings and text transcriptions in a way that the phones, words and audio signals are time aligned. The result can be used in frequently used software such as praat to go directly to a sound section of interest.

Especially relevant for

everybody who has audio signals and transcriptions, for example researchers working with:

  • linguistics
  • phonetics
  • anthropology
  • ethnology
  • media studies
  • educational research
  • conversation analysis
  • speech pathology
  • political science
  • speech technology

Starting point:

a recording of a German sentence (*.wav) and the orthographic transcript (*.txt)

Task:

a hierarchical segmentation in words, phonology, phones in praat compatible format (*.TextGrid); phonetic segments should be encoded in SAM-PA

Solution:

Web Interface BAS WebMAUS Basic

Related CLARIN-D tools and services

  • WebMAUS-Pipeline
  • WebMAUS-Basic
  • From text to phonological pronunciation

Short guide on how to use WebMAUS Basic

Preparation:

  1. download the ZIP package ftp://ftp.phonetik.uni-muenchen.de/pub/BAS/CLARIN/USE_CASES/multipleFileSegmentation.zip and up-pack it onto your local desktop folder; there should be a directory called 'examples-WebMAUS' on your desktop.

Web Interface:

  1. goto http://clarin.phonetik.uni-muenchen.de/BASWebServices
  2. select webservice 'WebMAUS Basic'
  3. click on the button 'Signal-file-upload' and select the file 'example2.wav' from the directory 'examples-WebMAUS' for upload
  4. click on the button 'Text-file-upload' and select the file 'example2.txt' from the directory 'examples-WebMAUS' for upload. You can see the text and listen to the uploaded signal
  5. execute WebMAUS with the following option
    • Language = German
    • Confirm the terms-of-usage and press the button 'Run Web Service'
  6. after completion the resulting TextGrid file is shown below; click on "Save result" to download it to your local system, or you can click on the button 'Segmentation Preview' below to open a new window in your browser and inspect the segmentation result directly using the EmuLabeller.
  7. Suggestion: mark a segment with the mouse and then press 'c' on your keyboard to play the sound.