Real-Time Joint Noise Suppression and Bandwidth Extension of Noisy Reverberant Wideband Speech

Artificially extending the bandwidth of speech in real-time applications that are band-limited to 16 kHz (known as wideband) or lower sample rates such as VoIP or communication over Bluetooth, can significantly improve its perceptual quality. Typically, dry clean speech is assumed as input to estimate the missing spectral information. However, such an assumption falls short if the input speech is reverberant or has been contaminated by noise, resulting in audible artifacts. We propose a real-time low-complexity multitasking neural network capable of performing noise suppression and bandwidth extension from 16 kHz to 48 kHz (fullband) on a CPU, preventing such issues even if the noise cannot be completely removed from the input. Instead of employing a monolithic model, we adopt a modular approach and complexity reduction methods that result in a more compact model than the sum of its parts while improving its performance.