mlsanityData to 14 Oct 2026

Visualizing Automatic Speech Recognition -- Means for a Better Understanding?

arXiv 2202.00673 ↗First posted 1 Feb 2022cs.LG cs.AI cs.CLcited by 6

Not on the map. Why?

Abstract (arXiv)

Automatic speech recognition (ASR) is improving ever more at mimicking human speech processing. The functioning of ASR, however, remains to a large extent obfuscated by the complex structure of the deep neural networks (DNNs) they are based on. In this paper, we show how so-called attribution methods, that we import from image recognition and suitably adapt to handle audio data, can help to clarify the working of ASR. Taking DeepSpeech, an end-to-end model for ASR, as a case study, we show how these techniques help to visualize which features of the input are the most influential in determining the output. We focus on three visualization techniques: Layer-wise Relevance Propagation (LRP), Saliency Maps, and Shapley Additive Explanations (SHAP). We compare these methods and discuss potential further applications, such as in the detection of adversarial examples.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 15 of its 25 references that are in the map. First posted on arXiv 1 Feb 2022.

Builds on

15 of 25 references.

References · 15

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  2. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  3. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  4. 2020Captum: A unified and generic model interpretability library for PyTorchcited by 547
  5. 2018Explaining Explanations: An Overview of Interpretability of Machine Learningcited by 508
  6. 2014Deep Speech: Scaling up end-to-end speech recognitioncited by 354
  7. 2017The (Un)reliability of saliency methodscited by 333
  8. 2018Audio Adversarial Examples: Targeted Attacks on Speech-to-Textcited by 274
  9. 2019Imperceptible, Robust, and Targeted Adversarial Examples for Automatic Speech Recognitioncited by 124
  10. 2021Do Feature Attribution Methods Correctly Attribute Features?cited by 98
  11. 2019On the Connection Between Adversarial Robustness and Saliency Map Interpretabilitycited by 80
  12. 2020SoK: The Faults in our ASRs: An Overview of Attacks against Automatic Speech Recognition and Speaker Identification Systemscited by 46
3 more
  1. 2019Saliency Methods for Explaining Adversarial Attackscited by 16
  2. 2019Towards a Unified Evaluation of Explanation Methods without Ground Truthcited by 6
  3. 2020Towards Resistant Audio Adversarial Examplescited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 3; 2026: 1.

6 papers.

  1. 2024SPES: Spectrogram Perturbation for Explainable Speech-to-Text Generationcited by 6
  2. 2025Voice, Bias, and Coreference: An Interpretability Study of Gender in Speech Translationcited by 5
  3. 2025The Unheard Alternative: Contrastive Explanations for Speech-to-Text Modelscited by 3
  4. 2024Using Shapley interactions to understand how models use structurecited by 1
  5. 2025Cross-Attention is Half Explanation in Speech-to-Text Modelscited by 1
  6. 2026Probing in the Wild: A Case Study of Self-Supervised Speech Representations on Mandarin Sub-dialects with Unsupervised Articulatory Analysiscited by 0