Traditional parametric coding of speech facilitates low rate but provides poor reconstruction quality because of the inadequacy of the model used. We describe how a WaveNet generative speech model can be used to generate high quality speech from the bit stream of a standard parametric coder operating at 2.4 kb/s. We compare this parametric coder with a waveform coder based on the same generative model and show that approximating the signal waveform incurs a large rate penalty. Our experiments confirm the high performance of the WaveNet based coder and show that the speech produced by the system is able to additionally perform implicit bandwidth extension and does not significantly impair recognition of the original speaker for the human listener, even when that speaker has not been used during the training of the generative model.

No abstract

In order to efficiently transmit and store speech signals, speech codecs create a minimally redundant representation of the input signal which is then decoded at the receiver with the best possible perceptual quality. In this work we demonstrate that a neural network architecture based on VQ-VAE with a WaveNet decoder can be used to perform very low bit-rate speech coding with high reconstruction quality. A prosody-transparent and speaker-independent model trained on the LibriSpeech corpus coding audio at 1.6 kbps exhibits perceptual quality which is around halfway between the MELP codec at 2.4 kbps and AMR-WB codec at 23.05 kbps. In addition, when training on high-quality recorded speech with the test speaker included in the training set, a model coding speech at 1.6 kbps produces output of similar perceptual quality to that generated by AMR-WB at 23.05 kbps.

To create systems that understand the sounds that humans are exposed to in everyday life, we need to represent sounds with features that can discriminate among many different sound classes. Here, we use a sound-ranking framework to quantitatively evaluate such representations in a large-scale task. We have adapted a machine-vision method, the passive-aggressive model for image retrieval (PAMIR), which efficiently learns a linear mapping from a very large sparse feature space to a large query-term space. Using this approach, we compare different auditory front ends and different ways of extracting sparse features from high-dimensional auditory images. We tested auditory models that use an adaptive pole-zero filter cascade (PZFC) auditory filter bank and sparse-code feature extraction from stabilized auditory images with multiple vector quantizers. In addition to auditory image models, we compare a family of more conventional mel-frequency cepstral coefficient (MFCC) front ends. The experimental results show a significant advantage for the auditory models over vector-quantized MFCCs. When thousands of sound files with a query vocabulary of thousands of words were ranked, the best precision at top-1 was 73% and the average precision was 35%, reflecting a 18% improvement over the best competing MFCC front end.

This paper investigates the theoretical basis for estimating vocal-tract length (VTL) from the formant frequencies of vowel sounds. A statistical inference model was developed to characterize the relationship between vowel type and VTL, on the one hand, and formant frequency and vocal cavity size, on the other. The model was applied to two well known developmental studies of formant frequency. The results show that VTL is the major source of variability after vowel type and that the contribution due to other factors like developmental changes in oral-pharyngeal ratio is small relative to the residual measurement noise. The results suggest that speakers adjust the shape of the vocal tract as they grow to maintain a specific pattern of formant frequencies for individual vowels. This formant-pattern hypothesis motivates development of a statisticalinference model for estimating VTL from formant-frequency data. The technique is illustrated using a third developmental study of formant frequencies. The VTLs of the speakers are estimated and used to provide a more accurate description of the complicated relationship between VTL and glottal pulse rate as children mature into adults.