# Prepending or Cross-Attention for Speech-to-Text? An Empirical Comparison

arXiv: https://arxiv.org/abs/2501.02370 · HTML: https://mlsanity.com/paper/2501.02370/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 49 of its 64 references that are in the map.
- First posted on arXiv 4 Jan 2025.

## Builds on (most cited first, 30 of 49)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [fairseq: A Fast, Extensible Toolkit for Sequence Modeling](https://mlsanity.com/paper/1904.01038/) (2019)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [Chameleon: Mixed-Modal Early-Fusion Foundation Models](https://mlsanity.com/paper/2405.09818/) (2024)
- [Sequence-Level Knowledge Distillation](https://mlsanity.com/paper/1606.07947/) (2016)
- [Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models](https://mlsanity.com/paper/2311.07919/) (2023)
- [Listen, Attend and Spell](https://mlsanity.com/paper/1508.01211/) (2015)
- [AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head](https://mlsanity.com/paper/2304.12995/) (2023)
- [SpeechT5: Unified-Modal Encoder-Decoder Pre-Training for Spoken Language Processing](https://mlsanity.com/paper/2110.07205/) (2021)
- [fairseq S2T: Fast Speech-to-Text Modeling with fairseq](https://mlsanity.com/paper/2010.05171/) (2020)
- [On decoder-only architecture for speech-to-text and large language model integration](https://mlsanity.com/paper/2307.03917/) (2023)
- [Prompting Large Language Models with Speech Recognition Abilities](https://mlsanity.com/paper/2307.11795/) (2023)
- [X-LLM: Bootstrapping Advanced Large Language Models by Treating Multi-Modalities as Foreign Languages](https://mlsanity.com/paper/2305.04160/) (2023)
- [Understanding Knowledge Distillation in Non-autoregressive Machine Translation](https://mlsanity.com/paper/1911.02727/) (2019)
- [Sequence-to-Sequence Models Can Directly Translate Foreign Speech](https://mlsanity.com/paper/1703.08581/) (2017)
- [Listen and Translate: A Proof of Concept for End-to-End Speech-to-Text Translation](https://mlsanity.com/paper/1612.01744/) (2016)
- [Speech Translation and the End-to-End Promise: Taking Stock of Where We Are](https://mlsanity.com/paper/2004.06358/) (2020)
- [SALM: Speech-augmented Language Model with In-context Learning for Speech Recognition and Translation](https://mlsanity.com/paper/2310.09424/) (2023)
- [SLM: Bridge the thin gap between speech and text foundation models](https://mlsanity.com/paper/2310.00230/) (2023)
- [End-to-End Automatic Speech Translation of Audiobooks](https://mlsanity.com/paper/1802.04200/) (2018)
- [STEMM: Self-learning with Speech-text Manifold Mixup for Speech Translation](https://mlsanity.com/paper/2203.10426/) (2022)
- [Bridging the Modality Gap for Speech-to-Text Translation](https://mlsanity.com/paper/2010.14920/) (2020)

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

- [Speech Language Models for Under-Represented Languages: Insights from Wolof](https://mlsanity.com/paper/2509.15362/) (2025)
- [Cross-lingual Matryoshka Representation Learning across Speech and Text](https://mlsanity.com/paper/2602.19991/) (2026)
- [FBK's Long-form SpeechLLMs for IWSLT 2026 Instruction Following](https://mlsanity.com/paper/2606.26819/) (2026)
- [COALA: Robust Contextualized Speech-augmented Language Modeling for ASR via Contrastive Regularizer and Biasing Score Estimation](https://mlsanity.com/paper/2607.08117/) (2026)
- [CAFE: Confining Feature Distillation to Language-Routed Adapters for Compact Multilingual Automatic Speech Recognition](https://mlsanity.com/paper/2610.19134/) (2026)

## Abstract (arXiv)

Following the remarkable success of Large Language Models (LLMs) in NLP tasks, there is increasing interest in extending their capabilities to speech -- the most common form of communication. The most widespread approach to integrating speech into LLMs is dense feature prepending (DFP), which prepends the projected speech representations to the textual representations, allowing end-to-end training with a speech encoder. This raises questions about the need for a sophisticated speech encoder for DFP and how its performance compares with a standard encoder-decoder (i.e., cross-attention) architecture. We compare DFP and cross-attention under a variety of configurations, such as CTC compression, sequence-level knowledge distillation, on monolingual, bilingual, and multilingual models. To perform a controlled architectural comparison, we train all models from scratch rather than using large pretrained models and use comparable data and parameter settings, testing speech-to-text recognition (ASR) and translation (ST) on MuST-C v1.0 and CoVoST2 datasets. Despite the wide adoption of DFP, our results do not indicate a clear advantage of DFP over cross-attention.