# LongSpeech: A Scalable Benchmark for Transcription, Translation and Understanding in Long Speech

arXiv: https://arxiv.org/abs/2601.13539 · HTML: https://mlsanity.com/paper/2601.13539/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jan 2026.

## Built on by (most cited first, 5 of 5)

- [Ontology Memory-Augmented ASR Correction for Long Text-Speech Interleaved Conversations](https://mlsanity.com/paper/2606.13464/) (2026)
- [MuLA-Bench: A Multilingual Long-Form Audio Understanding Benchmark via Multi-Tier Auditing](https://mlsanity.com/paper/2609.23416/) (2026)
- [WnW: Waxing-and-Waning KV Cache for Long-Form Speech LLMs](https://mlsanity.com/paper/2608.22704/) (2026)
- [agentic-ger: terminology recovery in long-form speech using global context](https://mlsanity.com/paper/2609.29428/) (2026)
- [Acoustic-to-Text KV Compression for Full-Duplex Speech Models](https://mlsanity.com/paper/2609.31224/) (2026)

## Abstract (arXiv)

Recent advances in audio-language models have demonstrated remarkable success on short, segment-level speech tasks. However, real-world applications such as meeting transcription, spoken document understanding, and conversational analysis require robust models capable of processing and reasoning over long-form audio. In this work, we present LongSpeech, a large-scale and scalable benchmark specifically designed to evaluate and advance the capabilities of speech models on long-duration audio. LongSpeech comprises over 100,000 speech segments, each approximately 10 minutes long, with rich annotations for ASR, speech translation, summarization, language detection, speaker counting, content separation, and question answering. We introduce a reproducible pipeline for constructing long-form speech benchmarks from diverse sources, enabling future extensions. Our initial experiments with state-of-the-art models reveal significant performance gaps, with models often specializing in one task at the expense of others and struggling with higher-level reasoning. These findings underscore the challenging nature of our benchmark. Our benchmark will be made publicly available to the research community.