# FastAdaSP: Multitask-Adapted Efficient Inference for Large Speech Language Model

arXiv: https://arxiv.org/abs/2410.03007 · HTML: https://mlsanity.com/paper/2410.03007/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Audio and omni-modal token compression, one of 12 directions in Efficient LLM inference and serving.
- Builds on 26 of its 39 references that are in the map.
- First posted on arXiv 3 Oct 2024.

## Placed in

- Efficient LLM inference and serving / [Audio and omni-modal token compression](https://mlsanity.com/inference/audio-and-omni-modal-token-compression/)

## Builds on (most cited first, 26 of 26)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversations](https://mlsanity.com/paper/1810.02508/) (2018)
- [Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models](https://mlsanity.com/paper/2311.07919/) (2023)
- [LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models](https://mlsanity.com/paper/2403.15388/) (2024)
- [Clotho: An Audio Captioning Dataset](https://mlsanity.com/paper/1910.09387/) (2019)
- [AudioPaLM: A Large Language Model That Can Speak and Listen](https://mlsanity.com/paper/2306.12925/) (2023)
- [Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Loss](https://mlsanity.com/paper/2002.02562/) (2020)
- [How2: A Large-scale Dataset for Multimodal Language Understanding](https://mlsanity.com/paper/1811.00347/) (2018)
- [MuTual: A Dataset for Multi-Turn Dialogue Reasoning](https://mlsanity.com/paper/2004.04494/) (2020)
- [LOOK-M: Look-Once Optimization in KV Cache for Efficient Multimodal Long-Context Inference](https://mlsanity.com/paper/2406.18139/) (2024)
- [E-Branchformer: Branchformer with Enhanced merging for speech recognition](https://mlsanity.com/paper/2210.00077/) (2022)
- [Get More with LESS: Synthesizing Recurrence with KV Cache Compression for Efficient LLM Inference](https://mlsanity.com/paper/2402.09398/) (2024)
- [Voxtlm: unified decoder-only models for consolidating speech recognition/synthesis and speech/text continuation tasks](https://mlsanity.com/paper/2309.07937/) (2023)
- [Whispering LLaMA: A Cross-Modal Generative Error Correction Framework for Speech Recognition](https://mlsanity.com/paper/2310.06434/) (2023)
- [SALMON: Self-Alignment with Instructable Reward Models](https://mlsanity.com/paper/2310.05910/) (2023)
- [Efficient conformer: Progressive downsampling and grouped attention for automatic speech recognition](https://mlsanity.com/paper/2109.01163/) (2021)
- [MLP Can Be A Good Transformer Learner](https://mlsanity.com/paper/2404.05657/) (2024)
- [Accelerating Transducers through Adjacent Token Merging](https://mlsanity.com/paper/2306.16009/) (2023)
- [SynesLM: A Unified Approach for Audio-visual Speech Recognition and Translation via Language Model and Synthetic Data](https://mlsanity.com/paper/2408.00624/) (2024)
- [Skipformer: A Skip-and-Recover Strategy for Efficient Speech Recognition](https://mlsanity.com/paper/2403.08258/) (2024)

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

- [AngelSlim: A more accessible, comprehensive, and efficient toolkit for large model compression](https://mlsanity.com/paper/2602.21233/) (2026)
- [FastLongSpeech: Enhancing Large Speech-Language Models for Efficient Long-Speech Processing](https://mlsanity.com/paper/2507.14815/) (2025)
- [Stride-k Subsampling: Train-Free Audio Token Reduction for Whisper](https://mlsanity.com/paper/2608.30927/) (2026)
- [Acoustic-to-Text KV Compression for Full-Duplex Speech Models](https://mlsanity.com/paper/2609.31224/) (2026)
- [Audio Token Attention Is Predictable Before the Language Model Runs](https://mlsanity.com/paper/2609.38878/) (2026)

## Abstract (arXiv)

In this study, we aim to explore Multitask Speech Language Model (SpeechLM) efficient inference via token reduction. Unlike other modalities such as vision or text, speech has unique temporal dependencies, making previous efficient inference works on other modalities not directly applicable. Furthermore, methods for efficient SpeechLM inference on long sequence and sparse signals remain largely unexplored. Then we propose FastAdaSP, a weighted token merging framework specifically designed for various speech-related tasks to improve the trade-off between efficiency and performance. Experimental results on WavLLM and Qwen-Audio show that our method achieves the state-of-the-art (SOTA) efficiency-performance trade-off compared with other baseline methods. Specifically, FastAdaSP achieved 7x memory efficiency and 1.83x decoding throughput without any degradation on tasks like Emotion Recognition (ER) and Spoken Question Answering (SQA). The code will be available at https://github.com/yichen14/FastAdaSP