# Listening or Reading? Evaluating Speech Awareness in Chain-of-Thought Speech-to-Text Translation

arXiv: https://arxiv.org/abs/2510.03115 · HTML: https://mlsanity.com/paper/2510.03115/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Audio-language model reasoning, one of 17 directions in LLM reasoning.
- Builds on 21 of its 28 references that are in the map.
- First posted on arXiv 3 Oct 2025.

## Placed in

- LLM reasoning / [Audio-language model reasoning](https://mlsanity.com/lr/audio-language-model-reasoning/)

## Builds on (most cited first, 21 of 21)

- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities](https://mlsanity.com/paper/2305.11000/) (2023)
- [MLS: A Large-Scale Multilingual Dataset for Speech Research](https://mlsanity.com/paper/2012.03411/) (2020)
- [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)
- [AudioPaLM: A Large Language Model That Can Speak and Listen](https://mlsanity.com/paper/2306.12925/) (2023)
- [xCOMET: Transparent Machine Translation Evaluation through Fine-grained Error Detection](https://mlsanity.com/paper/2310.10482/) (2023)
- [Liger Kernel: Efficient Triton Kernels for LLM Training](https://mlsanity.com/paper/2410.10989/) (2024)
- [Europarl-ST: A Multilingual Corpus For Speech Translation Of Parliamentary Debates](https://mlsanity.com/paper/1911.03167/) (2019)
- [Listen and Translate: A Proof of Concept for End-to-End Speech-to-Text Translation](https://mlsanity.com/paper/1612.01744/) (2016)
- [Textually Pretrained Speech Language Models](https://mlsanity.com/paper/2305.13009/) (2023)
- [Inseq: An Interpretability Toolkit for Sequence Generation Models](https://mlsanity.com/paper/2302.13942/) (2023)
- [Speech Translation and the End-to-End Promise: Taking Stock of Where We Are](https://mlsanity.com/paper/2004.06358/) (2020)
- [Quantifying Context Mixing in Transformers](https://mlsanity.com/paper/2301.12971/) (2023)
- [Speech Translation with Large Language Models: An Industrial Practice](https://mlsanity.com/paper/2312.13585/) (2023)
- [Chain-of-Thought Prompting for Speech Translation](https://mlsanity.com/paper/2409.11538/) (2024)
- [Speech is More Than Words: Do Speech-to-Text Translation Systems Leverage Prosody?](https://mlsanity.com/paper/2410.24019/) (2024)
- [Salamandra Technical Report](https://mlsanity.com/paper/2502.08489/) (2025)
- [From TOWER to SPIRE: Adding the Speech Modality to a Translation-Specialist LLM](https://mlsanity.com/paper/2503.10620/) (2025)
- [Speech-to-Text Translation with Phoneme-Augmented CoT: Enhancing Cross-Lingual Transfer in Low-Resource Scenarios](https://mlsanity.com/paper/2505.24691/) (2025)
- [From SALAMANDRA to SALAMANDRATA: BSC Submission for WMT25 General Machine Translation Shared Task](https://mlsanity.com/paper/2508.12774/) (2025)

## Built on by (most cited first, 1 of 1)

- [Transcribe, Translate, and Optimize: Joint Reward Learning for Speech Translation](https://mlsanity.com/paper/2609.26536/) (2026)

## Abstract (arXiv)

Speech-to-Text Translation (S2TT) systems built from Automatic Speech Recognition (ASR) and Text-to-Text Translation (T2TT) modules face two major limitations: error propagation and the inability to exploit prosodic or other acoustic cues. Chain-of-Thought (CoT) prompting has recently been introduced, with the expectation that jointly accessing speech and transcription will overcome these issues. Analyzing CoT through attribution methods, robustness evaluations with corrupted transcripts, and prosody-awareness, we find that it largely mirrors cascaded behavior, relying mainly on transcripts while barely leveraging speech. Simple training interventions, such as adding Direct S2TT data or noisy transcript injection, enhance robustness and increase speech attribution. These findings challenge the assumed advantages of CoT and highlight the need for architectures that explicitly integrate acoustic information into translation.