Deep Learning-Based Single-Ended Objective Quality Measures for Time-Scale Modified Audio
This provides a practical tool for audio engineers and researchers to assess TSM algorithms without needing reference signals, though it is incremental as it builds on existing datasets and methods.
The paper tackles the problem of objectively evaluating audio quality after Time-Scale Modification (TSM) by proposing two single-ended measures that do not require a reference signal, achieving average Root Mean Squared Errors of 0.608 and 0.576 and mean Pearson correlations of 0.771 and 0.794.
Objective evaluation of audio processed with Time-Scale Modification (TSM) is seeing a resurgence of interest. Recently, a labelled time-scaled audio dataset was used to train an objective measure for TSM evaluation. This DE measure was an extension of Perceptual Evaluation of Audio Quality, and required reference and test signals. In this paper, two single-ended objective quality measures for time-scaled audio are proposed that do not require a reference signal. Data driven features are created by either a convolutional neural network (CNN) or a bidirectional gated recurrent unit (BGRU) network and fed to a fully-connected network to predict subjective mean opinion scores. The proposed CNN and BGRU measures achieve an average Root Mean Squared Error of 0.608 and 0.576, and a mean Pearson correlation of 0.771 and 0.794, respectively. The proposed measures are used to evaluate TSM algorithms, and comparisons are provided for 16 TSM implementations. The objective measure is available at https://www.github.com/zygurt/TSM.