# SoundStream: An End-to-End Neural Audio Codec

arXiv: https://arxiv.org/abs/2107.03312 · HTML: https://mlsanity.com/paper/2107.03312/

## Key facts

- Cited by 771 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 60 references that are in the map.
- First posted on arXiv 7 Jul 2021.

## Builds on (most cited first, 30 of 34)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [WaveNet: A Generative Model for Raw Audio](https://mlsanity.com/paper/1609.03499/) (2016)
- [Generating Diverse High-Fidelity Images with VQ-VAE-2](https://mlsanity.com/paper/1906.00446/) (2019)
- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)](https://mlsanity.com/paper/1511.07289/) (2015)
- [LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speech](https://mlsanity.com/paper/1904.02882/) (2019)
- [The Perception-Distortion Tradeoff](https://mlsanity.com/paper/1711.06077/) (2017)
- [Jukebox: A Generative Model for Music](https://mlsanity.com/paper/2005.00341/) (2020)
- [MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesis](https://mlsanity.com/paper/1910.06711/) (2019)
- [High-Fidelity Generative Image Compression](https://mlsanity.com/paper/2006.09965/) (2020)
- [Speech Resynthesis from Discrete Disentangled Self-Supervised Representations](https://mlsanity.com/paper/2104.00355/) (2021)
- [Efficient Neural Audio Synthesis](https://mlsanity.com/paper/1802.08435/) (2018)
- [Parallel WaveNet: Fast High-Fidelity Speech Synthesis](https://mlsanity.com/paper/1711.10433/) (2017)
- [DDSP: Differentiable Digital Signal Processing](https://mlsanity.com/paper/2001.04643/) (2020)
- [SEGAN: Speech Enhancement Generative Adversarial Network](https://mlsanity.com/paper/1703.09452/) (2017)
- [SampleRNN: An Unconditional End-to-End Neural Audio Generation Model](https://mlsanity.com/paper/1612.07837/) (2016)
- [LPCNet: Improving Neural Speech Synthesis Through Linear Prediction](https://mlsanity.com/paper/1810.11846/) (2018)
- [ViSQOL v3: An Open Source Production Ready Objective Speech and Audio Metric](https://mlsanity.com/paper/2004.09584/) (2020)
- [The challenge of realistic music generation: modelling raw audio at scale](https://mlsanity.com/paper/1806.10474/) (2018)
- [Low Bit-Rate Speech Coding with VQ-VAE and a WaveNet Decoder](https://mlsanity.com/paper/1910.06464/) (2019)
- [A Wavenet for Speech Denoising](https://mlsanity.com/paper/1706.07162/) (2017)
- [A Spectral Energy Distance for Parallel Speech Synthesis](https://mlsanity.com/paper/2008.01160/) (2020)
- [SEANet: A Multi-modal Speech Enhancement Network](https://mlsanity.com/paper/2009.02095/) (2020)
- [Wavenet based low rate speech coding](https://mlsanity.com/paper/1712.01120/) (2017)
- [Speech Denoising with Deep Feature Losses](https://mlsanity.com/paper/1806.10522/) (2018)
- [End-to-End Optimized Speech Coding with Deep Neural Networks](https://mlsanity.com/paper/1710.09064/) (2017)
- [Real-time Speech Frequency Bandwidth Extension](https://mlsanity.com/paper/2010.10677/) (2020)
- [Exploring Speech Enhancement with Generative Adversarial Networks for Robust Speech Recognition](https://mlsanity.com/paper/1711.05747/) (2017)

## Built on by (most cited first, 30 of 771)

- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [High Fidelity Neural Audio Compression](https://mlsanity.com/paper/2210.13438/) (2022)
- [FAST: Efficient Action Tokenization for Vision-Language-Action Models](https://mlsanity.com/paper/2501.09747/) (2025)
- [AudioLM: a Language Modeling Approach to Audio Generation](https://mlsanity.com/paper/2209.03143/) (2022)
- [Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models](https://mlsanity.com/paper/2311.07919/) (2023)
- [VideoPoet: A Large Language Model for Zero-Shot Video Generation](https://mlsanity.com/paper/2312.14125/) (2023)
- [Simple and Controllable Music Generation](https://mlsanity.com/paper/2306.05284/) (2023)
- [MusicLM: Generating Music From Text](https://mlsanity.com/paper/2301.11325/) (2023)
- [Finite Scalar Quantization: VQ-VAE Made Simple](https://mlsanity.com/paper/2309.15505/) (2023)
- [High-Fidelity Audio Compression with Improved RVQGAN](https://mlsanity.com/paper/2306.06546/) (2023)
- [MoMask: Generative Masked Modeling of 3D Human Motions](https://mlsanity.com/paper/2312.00063/) (2023)
- [Voicebox: Text-Guided Multilingual Universal Speech Generation at Scale](https://mlsanity.com/paper/2306.15687/) (2023)
- [Compact 3D Gaussian Representation for Radiance Field](https://mlsanity.com/paper/2311.13681/) (2023)
- [F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching](https://mlsanity.com/paper/2410.06885/) (2024)
- [AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining](https://mlsanity.com/paper/2308.05734/) (2023)
- [AudioPaLM: A Large Language Model That Can Speak and Listen](https://mlsanity.com/paper/2306.12925/) (2023)
- [Make-An-Audio: Text-To-Audio Generation with Prompt-Enhanced Diffusion Models](https://mlsanity.com/paper/2301.12661/) (2023)
- [AudioGen: Textually Guided Audio Generation](https://mlsanity.com/paper/2209.15352/) (2022)
- [CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models](https://mlsanity.com/paper/2412.10117/) (2024)
- [AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head](https://mlsanity.com/paper/2304.12995/) (2023)
- [BigVGAN: A Universal Neural Vocoder with Large-Scale Training](https://mlsanity.com/paper/2206.04658/) (2022)
- [NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers](https://mlsanity.com/paper/2304.09116/) (2023)
- [NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion Models](https://mlsanity.com/paper/2403.03100/) (2024)
- [AnyGPT: Unified Multimodal LLM with Discrete Sequence Modeling](https://mlsanity.com/paper/2402.12226/) (2024)
- [Recommender Systems with Generative Retrieval](https://mlsanity.com/paper/2305.05065/) (2023)
- [MERT: Acoustic Music Understanding Model with Large-Scale Self-supervised Training](https://mlsanity.com/paper/2306.00107/) (2023)
- [Beyond Transmitting Bits: Context, Semantics, and Task-Oriented Communications](https://mlsanity.com/paper/2207.09353/) (2022)
- [Speak Foreign Languages with Your Own Voice: Cross-Lingual Neural Codec Language Modeling](https://mlsanity.com/paper/2303.03926/) (2023)
- [MaskGCT: Zero-Shot Text-to-Speech with Masked Generative Codec Transformer](https://mlsanity.com/paper/2409.00750/) (2024)
- [Behavior Generation with Latent Actions](https://mlsanity.com/paper/2403.03181/) (2024)

## Abstract (arXiv)

We present SoundStream, a novel neural audio codec that can efficiently compress speech, music and general audio at bitrates normally targeted by speech-tailored codecs. SoundStream relies on a model architecture composed by a fully convolutional encoder/decoder network and a residual vector quantizer, which are trained jointly end-to-end. Training leverages recent advances in text-to-speech and speech enhancement, which combine adversarial and reconstruction losses to allow the generation of high-quality audio content from quantized embeddings. By training with structured dropout applied to quantizer layers, a single model can operate across variable bitrates from 3kbps to 18kbps, with a negligible quality loss when compared with models trained at fixed bitrates. In addition, the model is amenable to a low latency implementation, which supports streamable inference and runs in real time on a smartphone CPU. In subjective evaluations using audio at 24kHz sampling rate, SoundStream at 3kbps outperforms Opus at 12kbps and approaches EVS at 9.6kbps. Moreover, we are able to perform joint compression and enhancement either at the encoder or at the decoder side with no additional latency, which we demonstrate through background noise suppression for speech.