# DarkStream: real-time speech anonymization with low latency

arXiv: https://arxiv.org/abs/2509.04667 · HTML: https://mlsanity.com/paper/2509.04667/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 4 Sep 2025.

## Builds on (most cited first, 20 of 20)

- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [High-Fidelity Audio Compression with Improved RVQGAN](https://mlsanity.com/paper/2306.06546/) (2023)
- [SpeechBrain: A General-Purpose Speech Toolkit](https://mlsanity.com/paper/2106.04624/) (2021)
- [Hierarchical Generative Modeling for Controllable Speech Synthesis](https://mlsanity.com/paper/1810.07217/) (2018)
- [Fish-Speech: Leveraging Large Language Models for Advanced Multilingual Text-to-Speech Synthesis](https://mlsanity.com/paper/2411.01156/) (2024)
- [Introducing the VoicePrivacy Initiative](https://mlsanity.com/paper/2005.01387/) (2020)
- [The VoicePrivacy 2024 Challenge Evaluation Plan](https://mlsanity.com/paper/2404.02677/) (2024)
- [Speaker anonymisation using the McAdams coefficient](https://mlsanity.com/paper/2011.01130/) (2020)
- [Design Choices for X-vector Based Speaker Anonymization](https://mlsanity.com/paper/2005.08601/) (2020)
- [Anonymizing Speech with Generative Adversarial Networks to Preserve Speaker Privacy](https://mlsanity.com/paper/2210.07002/) (2022)
- [Speaker anonymization using orthogonal Householder neural network](https://mlsanity.com/paper/2305.18823/) (2023)
- [Improving speaker de-identification with functional data analysis of f0 trajectories](https://mlsanity.com/paper/2203.16738/) (2022)
- [Are disentangled representations all you need to build speaker anonymization systems?](https://mlsanity.com/paper/2208.10497/) (2022)
- [End-to-end streaming model for low-latency speech anonymization](https://mlsanity.com/paper/2406.09277/) (2024)
- [VoiceMe: Personalized voice generation in TTS](https://mlsanity.com/paper/2203.15379/) (2022)
- [Distinctive and Natural Speaker Anonymization via Singular Value Transformation-assisted Matrix](https://mlsanity.com/paper/2405.10786/) (2024)

## Built on by (most cited first, 7 of 7)

- [Stream-Voice-Anon: Enhancing Utility of Real-Time Speaker Anonymization via Neural Audio Codec and Language Models](https://mlsanity.com/paper/2601.13948/) (2026)
- [TVTSyn: Content-Synchronous Time-Varying Timbre for Streaming Voice Conversion and Anonymization](https://mlsanity.com/paper/2602.09389/) (2026)
- [StreamVoiceAnon+: Emotion-Preserving Streaming Speaker Anonymization via Frame-Level Acoustic Distillation](https://mlsanity.com/paper/2603.06079/) (2026)
- [Privacy-Preserving End-to-End Full-Duplex Speech Dialogue Models](https://mlsanity.com/paper/2603.08179/) (2026)
- [PHONOS: PHOnetic Neutralization for Online Streaming Applications](https://mlsanity.com/paper/2603.27001/) (2026)
- [VOSSA: Voiceprint Optimization for Streaming Speech Architectures](https://mlsanity.com/paper/2609.38887/) (2026)
- [LiveAnon: Improving Privacy and Utility in Streaming Speaker Anonymization with Codec Language Models](https://mlsanity.com/paper/2610.15379/) (2026)

## Abstract (arXiv)

We propose DarkStream, a streaming speech synthesis model for real-time speaker anonymization. To improve content encoding under strict latency constraints, DarkStream combines a causal waveform encoder, a short lookahead buffer, and transformer-based contextual layers. To further reduce inference time, the model generates waveforms directly via a neural vocoder, thus removing intermediate mel-spectrogram conversions. Finally, DarkStream anonymizes speaker identity by injecting a GAN-generated pseudo-speaker embedding into linguistic features from the content encoder. Evaluations show our model achieves strong anonymization, yielding close to 50% speaker verification EER (near-chance performance) on the lazy-informed attack scenario, while maintaining acceptable linguistic intelligibility (WER within 9%). By balancing low-latency, robust privacy, and minimal intelligibility degradation, DarkStream provides a practical solution for privacy-preserving real-time speech communication.