Sommelier: Scalable Open Multi-turn Audio Pre-processing for Full-duplex Speech Language Models
This addresses the problem of enabling real-time, natural human-computer interaction for AI developers, but appears incremental as it focuses on improving data processing rather than introducing a new model or paradigm.
The paper tackles the scarcity of high-quality multi-speaker conversational data for full-duplex speech language models by presenting a robust and scalable open-source data processing pipeline, though no concrete results or numbers are provided.
As the paradigm of AI shifts from text-based LLMs to Speech Language Models (SLMs), there is a growing demand for full-duplex systems capable of real-time, natural human-computer interaction. However, the development of such models is constrained by the scarcity of high-quality, multi-speaker conversational data, as existing large-scale resources are predominantly single-speaker or limited in volume. Addressing the complex dynamics of natural dialogue, such as overlapping and back-channeling remains a challenge, with standard processing pipelines suffering from diarization errors and ASR hallucinations. To bridge this gap, we present a robust and scalable open-source data processing pipeline designed for full-duplex model.