Rapping-Singing Voice Synthesis based on Phoneme-level Prosody Control

TL;DR

This paper introduces a neural TTS system capable of synthesizing high-quality rapping and singing voices with phoneme-level prosody control, adaptable to any speaker using limited data and enhanced with traditional DSP techniques.

Contribution

It presents a novel Tacotron-based multispeaker model with phoneme-level prosody control, fine-tuning for unseen speakers, and integration of DSP algorithms for improved singing voice synthesis.

Findings

High-quality singing voice synthesis achieved

Enhanced naturalness over existing systems

Effective adaptation to new speakers with limited data

Abstract

In this paper, a text-to-rapping/singing system is introduced, which can be adapted to any speaker's voice. It utilizes a Tacotron-based multispeaker acoustic model trained on read-only speech data and which provides prosody control at the phoneme level. Dataset augmentation and additional prosody manipulation based on traditional DSP algorithms are also investigated. The neural TTS model is fine-tuned to an unseen speaker's limited recordings, allowing rapping/singing synthesis with the target's speaker voice. The detailed pipeline of the system is described, which includes the extraction of the target pitch and duration values from an a capella song and their conversion into target speaker's valid range of notes before synthesis. An additional stage of prosodic manipulation of the output via WSOLA is also investigated for better matching the target duration values. The synthesized utterances can be mixed with an instrumental accompaniment track to produce a complete song. The proposed system is evaluated via subjective listening tests as well as in comparison to an available alternate system which also aims to produce synthetic singing voice from read-only training data. Results show that the proposed approach can produce high quality rapping/singing voice with increased naturalness.