Speaking Speed Control of End-to-End Speech Synthesis using Sentence-Level Conditioning

TL;DR

This paper introduces a controllable end-to-end speech synthesis system that adjusts speaking speed at the sentence level using a speed ratio, maintaining speech quality and attributes without external models.

Contribution

It presents a novel sentence-level speed control method for end-to-end TTS that preserves speech naturalness and attributes, unlike previous duration-based approaches.

Findings

Generated speech at various speeds was more natural than existing methods.

The system effectively controls speaking speed while maintaining pitch and style.

Listening tests confirmed improved naturalness, especially at slow speeds.

Abstract

This paper proposes a controllable end-to-end text-to-speech (TTS) system to control the speaking speed (speed-controllable TTS; SCTTS) of synthesized speech with sentence-level speaking-rate value as an additional input. The speaking-rate value, the ratio of the number of input phonemes to the length of input speech, is adopted in the proposed system to control the speaking speed. Furthermore, the proposed SCTTS system can control the speaking speed while retaining other speech attributes, such as the pitch, by adopting the global style token-based style encoder. The proposed SCTTS does not require any additional well-trained model or an external speech database to extract phoneme-level duration information and can be trained in an end-to-end manner. In addition, our listening tests on fast-, normal-, and slow-speed speech showed that the SCTTS can generate more natural speech than other phoneme duration control approaches which increase or decrease duration at the same rate for the entire sentence, especially in the case of slow-speed speech.