Reduction of Subjective Listening Effort for TV Broadcast Signals with Recurrent Neural Networks

TL;DR

This paper presents a speech enhancement system using recurrent neural networks to separate speech from background noise in TV broadcasts, reducing listening effort and improving perceived sound quality for listeners.

Contribution

It introduces a novel RNN-based approach that separates and remixes speech signals at higher SNR, enhancing audio clarity in broadcast signals.

Findings

Reduces listening effort by around 2 points on a 13-point scale

Increases perceived sound quality compared to original mixture

Effective in real TV-broadcast scenarios

Abstract

Listening to the audio of TV broadcast signals can be challenging for hearing-impaired as well as normal-hearing listeners, especially when background sounds are prominent or too loud compared to the speech signal. This can result in a reduced satisfaction and increased listening effort of the listeners. Since the broadcast sound is usually premixed, we perform a subjective evaluation for quantifying the potential of speech enhancement systems based on audio source separation and recurrent neural networks (RNN). Recently, RNNs have shown promising results in the context of sound source separation and real-time signal processing. In this paper, we separate the speech from the background signals and remix the separated sounds at a higher signal-to-noise ratio. This differs from classic speech enhancement, where usually only the extracted speech signal is exploited. The subjective evaluation with 20 normal-hearing subjects on real TV-broadcast material shows that our proposed enhancement system is able to reduce the listening effort by around 2 points on a 13-point listening effort rating scale and increases the perceived sound quality compared to the original mixture.