# Echoes of Phonetics: Unveiling Relevant Acoustic Cues for ASR via Feature Attribution

arXiv: https://arxiv.org/abs/2506.02181 · HTML: https://mlsanity.com/paper/2506.02181/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 38 references that are in the map.
- First posted on arXiv 2 Jun 2025.

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

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [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)
- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation](https://mlsanity.com/paper/2101.00390/) (2021)
- [TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptation](https://mlsanity.com/paper/1805.04699/) (2018)
- [fairseq S2T: Fast Speech-to-Text Modeling with fairseq](https://mlsanity.com/paper/2010.05171/) (2020)
- [Self-Supervised Speech Representations are More Phonetic than Semantic](https://mlsanity.com/paper/2406.08619/) (2024)
- [Probing self-supervised speech models for phonetic and phonemic information: a case study in aspiration](https://mlsanity.com/paper/2306.06232/) (2023)
- [Probing phoneme, language and speaker information in unsupervised speech representations](https://mlsanity.com/paper/2203.16193/) (2022)
- [Encoding of phonology in a recurrent neural model of grounded speech](https://mlsanity.com/paper/1706.03815/) (2017)
- [Twists, Humps, and Pebbles: Multilingual Speech Recognition Models Exhibit Gender Performance Gaps](https://mlsanity.com/paper/2402.17954/) (2024)
- [Interpretable Convolutional Filters with SincNet](https://mlsanity.com/paper/1811.09725/) (2018)
- [Analyzing Phonetic and Graphemic Representations in End-to-End Automatic Speech Recognition](https://mlsanity.com/paper/1907.04224/) (2019)
- [SPES: Spectrogram Perturbation for Explainable Speech-to-Text Generation](https://mlsanity.com/paper/2411.01710/) (2024)
- [Learning weakly supervised multimodal phoneme embeddings](https://mlsanity.com/paper/1704.06913/) (2017)
- [Gradient-Adjusted Neuron Activation Profiles for Comprehensive Introspection of Convolutional Speech Recognition Models](https://mlsanity.com/paper/2002.08125/) (2020)

## Built on by (most cited first, 2 of 2)

- [Listening with Attention: Entropy-Guided Explainability for Transformer-Based Audio Models](https://mlsanity.com/paper/2606.14647/) (2026)
- [Probing in the Wild: A Case Study of Self-Supervised Speech Representations on Mandarin Sub-dialects with Unsupervised Articulatory Analysis](https://mlsanity.com/paper/2606.25459/) (2026)

## Abstract (arXiv)

Despite significant advances in ASR, the specific acoustic cues models rely on remain unclear. Prior studies have examined such cues on a limited set of phonemes and outdated models. In this work, we apply a feature attribution technique to identify the relevant acoustic cues for a modern Conformer-based ASR system. By analyzing plosives, fricatives, and vowels, we assess how feature attributions align with their acoustic properties in the time and frequency domains, also essential for human speech perception. Our findings show that the ASR model relies on vowels' full time spans, particularly their first two formants, with greater saliency in male speech. It also better captures the spectral characteristics of sibilant fricatives than non-sibilants and prioritizes the release phase in plosives, especially burst characteristics. These insights enhance the interpretability of ASR models and highlight areas for future research to uncover potential gaps in model robustness.