Deep Feature CycleGANs: Speaker Identity Preserving Non-parallel Microphone-Telephone Domain Adaptation for Speaker Verification
(3 minutes introduction)
| Saurabh Kataria (Johns Hopkins University, USA), Jesús Villalba (Johns Hopkins University, USA), Piotr Żelasko (Johns Hopkins University, USA), Laureano Moro-Velázquez (Johns Hopkins University, USA), Najim Dehak (Johns Hopkins University, USA) |
|---|
With the increase in the availability of speech from varied domains, it is imperative to use such out-of-domain data to improve existing speech systems. Domain adaptation is a prominent pre-processing approach for this. We investigate it to adapt microphone speech to the telephone domain. Specifically, we explore CycleGAN-based unpaired translation of microphone data to improve the x-vector/speaker embedding network for Telephony Speaker Verification. We first demonstrate the efficacy of this on real challenging data and then, to improve further, we modify the CycleGAN formulation to make the adaptation task-specific. We modify CycleGAN’s identity loss, cycle-consistency loss, and adversarial loss to operate in the deep feature space. Deep features of a signal are extracted from an auxiliary (speaker embedding) network and, hence, preserves speaker identity. Our 3D convolution-based Deep Feature Discriminators (DFD) show relative improvements of 5–10% in terms of equal error rate. To dive deeper, we study a challenging scenario of pooling (adapted) microphone and telephone data with data augmentations and telephone codecs. Finally, we highlight the sensitivity of CycleGAN hyper-parameters and introduce a parameter called probability of adaptation.
