| title | GigaChat Audio: Time-aware Large Audio Language Model |
|---|---|
| authors | Aleksandr Kutsakov, Mariia Sadovina, Georgii Gospodinov, Alexandr Maximenko, Oleg Kutuzov, Pavel Bogomolov, Fyodor Minkin |
| date | 2026-07-11 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2607.10387 |
| affiliation | ai-sage |
| upvotes | 37 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Temporal grounding in long recordings remains challenging for audio-conditioned LLMs. We present a time-aware audio LLM that answers questions with explicit timestamps over up to 120 minutes of input. Our approach interleaves periodic time markers with continuous audio tokens using large-scale synthetic supervision from a cascaded pipeline. Our model achieves strong temporal-grounding accuracy on short and long benchmarks and supports time-anchored fragment descriptions and summaries. Extensive ablations examine how time representation, marker frequency, tokenization, and duration-mixture design affect accuracy and computational cost. We release model weights and datasets to support further research on time-aware audio understanding, available at https://huggingface.co/ai-sage/GigaChat3.1-Audio-10B-A1.8B. |
cite
@misc{kutsakov2026gigachat,
title={GigaChat Audio: Time-aware Large Audio Language Model},
author={Aleksandr Kutsakov and Mariia Sadovina and Georgii Gospodinov and Alexandr Maximenko and Oleg Kutuzov and Pavel Bogomolov and Fyodor Minkin},
year={2026},
eprint={2607.10387},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2607.10387}
}