Neural Source-Filter Architectures vs. Conventional Sound Modeling in Symbolic Music Synthesis: Perceptual Quality and Throughput

Speech synthesis and music audio generation from symbolic input differ in many aspects but share some similarities. In this study, we investigate how text-to-speech synthesis techniques can be used for piano MIDI-to-audio synthesis tasks. Our investigation includes Tacotron and neural source-filter waveform models as the basic components, with which we build MIDI-to-audio synthesis systems in similar ways to TTS frameworks. We also include reference systems using conventional sound modeling techniques such as sample-based and physical-modeling-based methods. The subjective experimental results

Research goal: How do neural source-filter architectures compare to conventional sound modeling techniques in terms of perceptual quality scores and generation throughput for symbolic music synthesis?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 7.9/10.