DISPLACE Corpus 2023: Dev Set
Shikha Baghel, Pratik Roy Chowdhuri, Somil Jain, Deepu Vijayasenan, Sriram Ganapathy · Zenodo (CERN European Organization for Nuclear Research) · 2023
The DISPLACE challenge entailed a first of kind task to perform speaker and language diarization on the same data, as the data contains multi-speaker social conversations in multi-lingual code-mixed speech. In multi-lingual communities, social conversations frequently involve code-mixed and code-switched speech. In such cases, various speech processing systems need to perform the speaker and language segmentation before any downstream task. The current speaker diarization systems are not equipped to handle multi-lingual conversations, while the language recognition systems may not be able to handle the same talker speaking in multiple languages within the same recording. With this motivation, the DISPLACE challenge addressed Speaker Diarization (SD) in multi-lingual settings and Language Diarization (LD) in multi-speaker settings, using the same underlying dataset. For this challenge, a natural multi-lingual, multi-speaker conversational dataset was distributed. More details about the dataset can be found at https://www.isca-speech.org/archive/pdfs/interspeech_2023/baghel23_interspeech.pdf