Benchmarking LLMs on the Massive Sound Embedding Benchmark (MSEB)
For researchers building audio AI systems, this work provides a benchmark comparison of LLM-based approaches versus specialized encoders, but conclusions are inconclusive.
This paper evaluates leading LLMs (Gemini, GPT) on the MSEB benchmark, finding a persistent modality gap between audio and text performance, with no clear optimal modeling approach.
The Massive Sound Embedding Benchmark (MSEB) has emerged as a standard for evaluating the functional breadth of audio models. While initial baselines focused on specialized encoders, the shift toward "audio-native" Large Language Models (LLMs) suggests a new paradigm where a single multimodal backbone may replace complex, task-specific pipelines. This paper provides a rigorous empirical evaluation of leading LLMs - including members from the Gemini and GPT families - across the eight core MSEB capabilities to assess their efficacy and audio-text parity. Our results indicate that while a significant modality gap persists regarding performance and robustness, the empirical evidence for an "optimal" modeling approach remains inconclusive. Ultimately, the choice between audionative and cascaded architectures depends heavily on specific use-case requirements and the underlying assumptions regarding latency, cost, and reasoning depth.