Speech-To-Singing Conversion In An Encoder-Decoder Framework

In this paper our goal is to convert a set of spoken lines into sung ones. Unlike previous signal processing based methods, we take a learning based approach to the problem. This allows us to automatically model various aspects of this transformation, thus overcoming dependence on specific inputs such as high quality singing templates or phoneme-score synchronization information. Specifically, we propose an encoder-decoder framework for our task. Given time-frequency representations of speech and a target melody we learn encodings that enable us to synthesize singing that preserves the linguistic content and timbre of the speaker while adhering to the target melody. We also propose a multi-task learning based objective to improve intelligibility. We present a thorough quantitative and qualitative analysis of our framework.