How to Annotate 100 Hours in 45 Minutes

Per Fallgren, Zofia Malisz, Jens Edlund

Speech data found in the wild hold many advantages over artificially constructed speech corpora in terms of ecological validity and cultural worth. Perhaps most importantly, there is a lot of it. However, the combination of great quantity, noisiness and variation poses a challenge for its access and processing. Generally speaking, automatic approaches to tackle the problem require good labels for training, while manual approaches require time. In this study, we provide further evidence for a semi-supervised, human-in-the-loop framework that previously has shown promising results for browsing and annotating large quantities of found audio data quickly. The findings of this study show that a 100-hour long subset of the Fearless Steps corpus can be annotated for speech activity in less than 45 minutes, a fraction of the time it would take traditional annotation methods, without a loss in performance.

 DOI: 10.21437/Interspeech.2019-1648

Cite as: Fallgren, P., Malisz, Z., Edlund, J. (2019) How to Annotate 100 Hours in 45 Minutes. Proc. Interspeech 2019, 341-345, DOI: 10.21437/Interspeech.2019-1648.

  author={Per Fallgren and Zofia Malisz and Jens Edlund},
  title={{How to Annotate 100 Hours in 45 Minutes}},
  booktitle={Proc. Interspeech 2019},