# G-STAR: End-to-End Global Speaker-Tracking Attributed Recognition

arXiv: https://arxiv.org/abs/2603.10468 · HTML: https://mlsanity.com/paper/2603.10468/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 31 references that are in the map.
- First posted on arXiv 11 Mar 2026.

## Builds on (most cited first, 25 of 25)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models](https://mlsanity.com/paper/2311.07919/) (2023)
- [SALMONN: Towards Generic Hearing Abilities for Large Language Models](https://mlsanity.com/paper/2310.13289/) (2023)
- [Qwen3-Omni Technical Report](https://mlsanity.com/paper/2509.17765/) (2025)
- [Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models](https://mlsanity.com/paper/2507.08128/) (2025)
- [Serialized Output Training for End-to-End Overlapped Speech Recognition](https://mlsanity.com/paper/2003.12687/) (2020)
- [Recent Advances in Speech Language Models: A Survey](https://mlsanity.com/paper/2410.03751/) (2024)
- [A Survey on Speech Large Language Models for Understanding](https://mlsanity.com/paper/2410.18908/) (2024)
- [AISHELL-4: An Open Source Dataset for Speech Enhancement, Separation, Recognition and Speaker Diarization in Conference Scenario](https://mlsanity.com/paper/2104.03603/) (2021)
- [Joint Speech Recognition and Speaker Diarization via Sequence Transduction](https://mlsanity.com/paper/1907.05337/) (2019)
- [FireRedASR: Open-Source Industrial-Grade Mandarin Speech Recognition Models from Encoder-Decoder to LLM Integration](https://mlsanity.com/paper/2501.14350/) (2025)
- [Streaming end-to-end multi-talker speech recognition](https://mlsanity.com/paper/2011.13148/) (2020)
- [Recognizing Overlapped Speech in Meetings: A Multichannel Separation Approach Using Neural Networks](https://mlsanity.com/paper/1810.03655/) (2018)
- [SpeakerLM: End-to-End Versatile Speaker Diarization and Recognition with Multimodal Large Language Models](https://mlsanity.com/paper/2508.06372/) (2025)
- [VIBEVOICE-ASR Technical Report](https://mlsanity.com/paper/2601.18184/) (2026)
- [Streaming Speaker-Attributed ASR with Token-Level Speaker Embeddings](https://mlsanity.com/paper/2203.16685/) (2022)
- [Summary on The Multilingual Conversational Speech Language Model Challenge: Datasets, Tasks, Baselines, and Methods](https://mlsanity.com/paper/2509.13785/) (2025)
- [Train Short, Infer Long: Speech-LLM Enables Zero-Shot Streamable Joint ASR and Diarization on Long Audio](https://mlsanity.com/paper/2511.16046/) (2025)
- [SURT 2.0: Advances in Transducer-based Multi-talker Speech Recognition](https://mlsanity.com/paper/2306.10559/) (2023)
- [Speaker Targeting via Self-Speaker Adaptation for Multi-talker ASR](https://mlsanity.com/paper/2506.22646/) (2025)
- [Streaming Sortformer: Speaker Cache-Based Online Speaker Diarization with Arrival-Time Ordering](https://mlsanity.com/paper/2507.18446/) (2025)
- [TagSpeech: End-to-End Multi-Speaker ASR and Diarization with Fine-Grained Temporal Grounding](https://mlsanity.com/paper/2601.06896/) (2026)
- [On Speaker Attribution with SURT](https://mlsanity.com/paper/2401.15676/) (2024)
- [MOSS Transcribe Diarize Technical Report](https://mlsanity.com/paper/2601.01554/) (2026)
- [WEST: LLM based Speech Toolkit for Speech Understanding, Generation, and Interaction](https://mlsanity.com/paper/2509.19902/) (2025)

## Abstract (arXiv)

We study timestamped speaker-attributed automatic speech recognition (SA-ASR) for long-form, multi-party speech with overlap. In this setting, chunk-wise inference must preserve meeting-level speaker identity consistency while producing time-stamped, speaker-labeled transcripts. Prior Speech-LLM systems tend to prioritize either local diarization or global labeling, lacking the ability to jointly model fine-grained temporal boundaries and robust cross-chunk identity linking. We propose G-STAR, an end-to-end framework that couples a cache-conditioned speaker-tracking module with a Speech-LLM transcription backbone. The tracker provides structured speaker cues with temporal grounding, and the LLM generates attributed text conditioned on these cues. G-STAR supports component-wise optimization and joint end-to-end training, enabling flexible learning under heterogeneous supervision and domain shift. Under chunk-wise decoding protocols, experiments on both oracle-segmented local evaluation and full-meeting global evaluation show strong speaker-attributed transcription performance.