# Teaching LLMs to Hear Who Spoke What: Metadata-Supervised Pretraining for Encoder-Free Speech-LLMs

arXiv: https://arxiv.org/abs/2610.01695 · HTML: https://mlsanity.com/paper/2610.01695/

## Key facts

- No citing papers in the map yet (data through 19 Oct 2026).
- Placed in Speech recognition models and training, one of 11 directions in Speech and audio language models.
- Builds on 28 of its 37 references that are in the map.
- First posted on arXiv 1 Oct 2026.

## Placed in

- Speech and audio language models / [Speech recognition models and training](https://mlsanity.com/speech/speech-recognition-models-and-training/)

## Builds on (most cited first, 28 of 28)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [VoxCeleb2: Deep Speaker Recognition](https://mlsanity.com/paper/1806.05622/) (2018)
- [SALMONN: Towards Generic Hearing Abilities for Large Language Models](https://mlsanity.com/paper/2310.13289/) (2023)
- [MLS: A Large-Scale Multilingual Dataset for Speech Research](https://mlsanity.com/paper/2012.03411/) (2020)
- [Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs](https://mlsanity.com/paper/2503.01743/) (2025)
- [FLEURS: Few-shot Learning Evaluation of Universal Representations of Speech](https://mlsanity.com/paper/2205.12446/) (2022)
- [VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation](https://mlsanity.com/paper/2101.00390/) (2021)
- [GigaSpeech: An Evolving, Multi-domain ASR Corpus with 10,000 Hours of Transcribed Audio](https://mlsanity.com/paper/2106.06909/) (2021)
- [Gemma 4 Technical Report](https://mlsanity.com/paper/2607.02770/) (2026)
- [TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptation](https://mlsanity.com/paper/1805.04699/) (2018)
- [CHiME-6 Challenge:Tackling Multispeaker Speech Recognition for Unsegmented Recordings](https://mlsanity.com/paper/2004.09249/) (2020)
- [Prompting Large Language Models with Speech Recognition Abilities](https://mlsanity.com/paper/2307.11795/) (2023)
- [WavLLM: Towards Robust and Adaptive Speech Large Language Model](https://mlsanity.com/paper/2404.00656/) (2024)
- [An Embarrassingly Simple Approach for LLM with Strong ASR Capacity](https://mlsanity.com/paper/2402.08846/) (2024)
- [Libriheavy: a 50,000 hours ASR corpus with punctuation casing and context](https://mlsanity.com/paper/2309.08105/) (2023)
- [Joint Audio and Speech Understanding](https://mlsanity.com/paper/2309.14405/) (2023)
- [SPGISpeech: 5,000 hours of transcribed financial audio for fully formatted end-to-end speech recognition](https://mlsanity.com/paper/2104.02014/) (2021)
- [Joint Speech Recognition and Speaker Diarization via Sequence Transduction](https://mlsanity.com/paper/1907.05337/) (2019)
- [Earnings-22: A Practical Benchmark for Accents in the Wild](https://mlsanity.com/paper/2203.15591/) (2022)
- [VIBEVOICE-ASR Technical Report](https://mlsanity.com/paper/2601.18184/) (2026)
- [Emilia: A Large-Scale, Extensive, Multilingual, and Diverse Dataset for Speech Generation](https://mlsanity.com/paper/2501.15907/) (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)
- [Advancing Multi-talker ASR Performance with Large Language Models](https://mlsanity.com/paper/2408.17431/) (2024)
- [LLM can Read Spectrogram: Encoder-free Speech-Language Modeling](https://mlsanity.com/paper/2606.10231/) (2026)

## Abstract (arXiv)

Encoder-based speech large language models (Speech-LLMs) commonly employ pretrained speech encoders that prioritize linguistic content but may discard fine-grained acoustic cues essential for speaker discrimination and paralinguistic understanding. Encoder-free Speech-LLMs instead map Mel-spectrogram features directly into the LLM input space through lightweight embedding layers, enabling the LLM to learn from low-level acoustic features. However, systematic pretraining strategies for encoder-free Speech-LLMs remain underexplored, limiting their ability to compensate for the absence of large-scale pretrained speech encoders. We propose metadata-supervised pretraining (MSP), which leverages speech attributes such as speaker identity and emotion to develop speaker-discriminative and paralinguistic capabilities. We further introduce speaker-aware utterance composition (SAUC) to strengthen speaker discrimination and apply random span masking to regularize pretraining. We primarily evaluate our approach on joint ASR and speaker diarization in multi-speaker conversations, complemented by experiments on paralinguistic speech-understanding tasks. Under matched training-data conditions, our encoder-free model outperforms its randomly initialized encoder-based counterpart. With limited metadata-annotated data, it is competitive with models using speech encoders pretrained on substantially larger corpora, outperforming them in several settings. These results demonstrate the potential of encoder-free architectures for building native multimodal LLMs that acquire diverse speech capabilities.