Streaming’s algorithms make it easy to avoid whole discographies – so in the interest of deeper listening, our writers dedicate time to the ones who might have got away ...

As we finish out 2025, we’ve been asking artists about their favorite albums of the year, and La Femme frontman and ...

Abstract: Neural waveform models have demonstrated better performance than conventional vocoders for statistical parametric speech synthesis. One of the best models, called WaveNet, uses an ...

Of all the brilliantly perverse moves Neil Young has made in his career, the 1982 release of Trans just may be the most brilliant. Not that the folks at Geffen, his label at the time, necessarily felt ...

Abstract: This paper introduces V2Coder, a non-autoregressive vocoder based on hierarchical variational autoencoders (VAEs). The hierarchical VAE with hierarchically extended prior and approximate ...