Retrieving Speaker Information from Personalized Acoustic Models for Speech Recognition
Not on the map. Why?
Abstract (arXiv)
The widespread of powerful personal devices capable of collecting voice of their users has opened the opportunity to build speaker adapted speech recognition system (ASR) or to participate to collaborative learning of ASR. In both cases, personalized acoustic models (AM), i.e. fine-tuned AM with specific speaker data, can be built. A question that naturally arises is whether the dissemination of personalized acoustic models can leak personal information. In this paper, we show that it is possible to retrieve the gender of the speaker, but also his identity, by just exploiting the weight matrix changes of a neural acoustic model locally adapted to this speaker. Incidentally we observe phenomena that may be useful towards explainability of deep neural networks in the context of speech processing. Gender can be identified almost surely using only the first layers and speaker verification performs well when using middle-up layers. Our experimental study on the TED-LIUM 3 dataset with HMM/TDNN models shows an accuracy of 95% for gender detection, and an Equal Error Rate of 9.07% for a speaker verification task by only exploiting the weights from personalized models that could be exchanged instead of user data.
Key facts. Cited by 2 papers in the map (data through 13 Oct 2026). Builds on 12 of its 17 references that are in the map. First posted on arXiv 7 Nov 2021.
Builds on
12 of 17 references.
References · 12
- 2020Three Approaches for Personalization with Applications to Federated Learningcited by 323
- 2018TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptationcited by 224
- 2018Federated Learning for Keyword Spottingcited by 95
- 2020Training Speech Recognition Models with Federated Learning: A Quality/Cost Frameworkcited by 46
- 2020Training Keyword Spotting Models on Non-IID Data with Federated Learningcited by 36
- 2017Analyzing Hidden Representations in End-to-End Automatic Speech Recognition Systemscited by 31
- 2021Federated Acoustic Modeling For Automatic Speech Recognitioncited by 20
- 2019The GDPR & Speech Data: Reflections of Legal and Technology Communities, First Steps towards a Common Understandingcited by 19
- 2019Analyzing Phonetic and Graphemic Representations in End-to-End Automatic Speech Recognitioncited by 16
- 2021Federated Learning in ASR: Not as Easy as You Thinkcited by 14
- 2019What does a network layer hear? Analyzing hidden representations of end-to-end ASR through speech synthesiscited by 13
- 2021Privacy attacks for automatic speech recognition acoustic models in a federated learning frameworkcited by 10
Built on by
Citing papers per year: 2023: 1; 2024: 1.
2 papers.