Cross-Corpus Data Augmentation for Acoustic Addressee Detection

no code implementations • WS 2019 • Oleg Akhtiamov, Ingo Siegert, Alexey Karpov, Wolfgang Minker

Mixup is shown to be beneficial for merging acoustic data (extracted features but not raw waveforms) from different domains that allows us to reach a higher classification performance on human-machine AD and also for training a multipurpose neural network that is capable of solving both human-machine and adult-child AD problems.

Cross-corpus Data Augmentation +1