Search papers, labs, and topics across Lattice.
VoXtream2 is introduced as a zero-shot full-stream TTS model enabling dynamic speaking-rate control and on-the-fly updates for interactive systems. It leverages a distribution matching mechanism over duration states combined with classifier-free guidance to enhance controllability and synthesis quality. The model achieves competitive performance on zero-shot benchmarks and operates at 4x real-time speed with low latency, despite its smaller size and reduced training data.
Control speaking rate on the fly in your TTS system with VoXtream2, which hits 4x real-time speeds and 74ms latency.
Full-stream text-to-speech (TTS) for interactive systems must start speaking with minimal delay while remaining controllable as text arrives incrementally. We present VoXtream2, a zero-shot full-stream TTS model with dynamic speaking-rate control that can be updated mid-utterance on the fly. VoXtream2 combines a distribution matching mechanism over duration states with classifier-free guidance across conditioning signals to improve controllability and synthesis quality. Prompt-text masking enables textless audio prompting, removing the need for prompt transcription. Across standard zero-shot benchmarks and a dedicated speaking-rate test set, VoXtream2 achieves competitive objective and subjective results against public baselines despite a smaller model and less training data. In full-stream mode, it runs 4 times faster than real time with 74 ms first-packet latency on a consumer GPU.
