Full-Reference Speech Quality Estimation with Attentional Siamese Neural Networks
This work addresses speech quality estimation for Voice-Over-IP networks, offering an incremental improvement by incorporating reference signals into neural network models.
The paper tackles the problem of full-reference speech quality prediction by introducing a siamese recurrent convolutional network with attention to align reference and degraded signals, achieving competitive performance on standard benchmarks.
In this paper, we present a full-reference speech quality prediction model with a deep learning approach. The model determines a feature representation of the reference and the degraded signal through a siamese recurrent convolutional network that shares the weights for both signals as input. The resulting features are then used to align the signals with an attention mechanism and are finally combined to estimate the overall speech quality. The proposed network architecture represents a simple solution for the time-alignment problem that occurs for speech signals transmitted through Voice-Over-IP networks and shows how the clean reference signal can be incorporated into speech quality models that are based on end-to-end trained neural networks.