Title | ||
---|---|---|
Mic2Mic: using cycle-consistent generative adversarial networks to overcome microphone variability in speech systems |
Abstract | ||
---|---|---|
Mobile and embedded devices are increasingly using microphones and audio-based computational models to infer user context. A major challenge in building systems that combine audio models with commodity microphones is to guarantee their accuracy and robustness in the real-world. Besides many environmental dynamics, a primary factor that impacts the robustness of audio models is microphone variability. In this work, we propose Mic2Mic - a machine-learned system component - which resides in the inference pipeline of audio models and at real-time reduces the variability in audio data caused by microphone-specific factors. Two key considerations for the design of Mic2Mic were: a) to decouple the problem of microphone variability from the audio task, and b) put minimal burden on end-users to provide training data. With these in mind, we apply the principles of cycle-consistent generative adversarial networks (CycleGANs) to learn Mic2Mic using unlabeled and unpaired data collected from different microphones. Our experiments show that Mic2Mic can recover between 66% to 89% of the accuracy lost due to microphone variability for two common audio tasks.
|
Year | DOI | Venue |
---|---|---|
2019 | 10.1145/3302506.3310398 | Proceedings of the 18th International Conference on Information Processing in Sensor Networks |
Keywords | Field | DocType |
GAN, microphone variability, robustness, speech models | Unpaired Data,Training set,Inference,Computer science,Robustness (computer science),Real-time computing,Speech recognition,Computational model,Generative grammar,Microphone,Adversarial system | Conference |
ISBN | Citations | PageRank |
978-1-4503-6284-9 | 1 | 0.39 |
References | Authors | |
18 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Akhil Mathur | 1 | 145 | 15.21 |
Anton Isopoussu | 2 | 2 | 2.08 |
Fahim Kawsar | 3 | 909 | 80.24 |
Nadia Berthouze | 4 | 123 | 14.38 |
Nicholas D. Lane | 5 | 4247 | 248.15 |