CRC 1412/2: Seemingly free (morpho)phonetic variation (SP C06)

We will reuse and enhance existing spoken corpora of German, one focus still being on non-native addressee register. The corpora are pre-processed and reformatted in order to achieve a unified collection that can be queried and analyzed. We will develop common annotation guidelines on speech and morphophonetic phenomena across corpora. We will find out the relation between phenomena and the different register-related parameters. Regarding modeling, we aim to extend existing addressee-based models with situational-functional parameters. After the annotation, the data will be re-integrated into existing data publications.