Bagpiper-TTS: Natural Language Guided Universal Speech Synthesis
This work addresses the need for flexible, user-friendly TTS systems that can handle diverse speech synthesis tasks from natural language instructions.
Bagpiper-TTS introduces a universal speech synthesis system that interprets natural language prompts to generate speech, achieving a 1.7% WER on Seed-TTS-Eval and matching dedicated models across multiple applications.
Classical TTS systems typically rely on rigid input formats and predefined metadata slots, limiting their ability to fulfill flexible user requirements. This paper introduces Bagpiper-TTS, a universal speech synthesis system that deals with diverse natural language user requests. Given a natural language prompt, Bagpiper-TTS first reasons over the users' intent to derive a rich caption, i.e., a comprehensive textual blueprint encompassing both transcription and nuanced metadata. Subsequently, this caption guides the synthesis of the target speech. Our model inherently supports a broad spectrum of tasks besides classical TTS applications, including multi-talker, intent-to-speech, role-play synthesis, singing voice synthesis, and more. Experimental results demonstrate that Bagpiper-TTS achieves an 1.7% Word Error Rate (WER) on the Seed-TTS-Eval benchmark and match the performance of dedicated models in both LLM-as-a-judge and human subjective evaluations across multiple applications.