S-PRESSO: Ultra Low Bitrate Sound Effect Compression With Diffusion Autoencoders And Offline Quantization

Zineb Lahrichi; Gaëtan Hadjeres; Gaël Richard; Geoffroy Peeters

cs.SDFeb 16, 2026

S-PRESSO: Ultra Low Bitrate Sound Effect Compression With Diffusion Autoencoders And Offline Quantization

Authors: Zineb Lahrichi, Gaëtan Hadjeres, Gaël Richard, Geoffroy Peeters

Organizations: IP Paris, SonyAI · Sony AI · LTCI, Télécom Paris, Institut Polytechnique de Paris · SonyAI ·

Abstract

Neural audio compression models have recently achieved extreme compression rates, enabling efficient latent generative modeling. Conversely, latent generative models have been applied to compression, pushing the limits of continuous and discrete approaches. However, existing methods remain constrained to low-resolution audio and degrade substantially at very low bitrates, where audible artifacts are prominent. In this paper, we present S-PRESSO, a 48kHz sound effect compression model that produces both continuous and discrete embeddings at ultra-low bitrates, down to 0.096 kbps, via offline quantization. Our model relies on a pretrained latent diffusion model to decode compressed audio embeddings learned by a latent encoder. Leveraging the generative priors of the diffusion decoder, we achieve extremely low frame rates, down to 1Hz (750x compression rate), producing convincing and realistic reconstructions at the cost of exact fidelity. Despite operating at high compression rates, we demonstrate that S-PRESSO outperforms both continuous and discrete baselines in audio quality, acoustic similarity and reconstruction metrics.

Figures & tables

Figure 1: Overview of our method. Step 1 : An audio clip is encoded into latent vectors $x_{0}$ by a low-compression audio autoencoder. It is then compressed into latents $z$ , which are upsampled by $f_{\phi}$ to condition the decoder $D_{\theta}$ , a DiT pretrained to reconstruct $x_{0}$ from noised inputs. $D_{\theta}$ is finetuned using LoRA adapters, jointly trained with the latent encoder $g_{\psi}$ and $f_{\phi}$ . Step 2 : The features $z$ are then quantized offline into $z_{q}$ . Step 3 : the diffusion decoder $D_{\theta}$ is finetuned on $z_{q}$ , to compensate for quantization-induced degradation.

Figure 2: (a) Overview of the latent encoder architecture (b) Conditioning mechanism within the diffusion decoder.

Method	Variant	$D$	Framerate	$R$	FAD $\downarrow$	FAD ${}_{\text{CLAP}}$ $\downarrow$	KAD ${}_{\text{CLAP}}$ $\downarrow$	CLAP ${}_{\text{audio}}$ $\uparrow$	Si-SDR $\uparrow$
AudioAE	–	128	100 Hz	4	0.008	0.008	0.15	0.90	22.3
StableAudio Open	–	64	21.5 Hz	32	0.78	0.066	1.25	0.78	0.48
S-PRESSO	$t=4$	64	25 Hz	30	0.48	0.038	0.57	0.76	3.21
Music2Latent	–	64	11 Hz	64	1.28	0.168	3.29	0.69	-10.5
S-PRESSO	$t=9$	64	11 Hz	68	0.59	0.050	0.77	0.76	-2.40
S-PRESSO	$t=20$	64	5 Hz	150	0.76	0.059	0.92	0.71	-8.80

Table 1: Performance in the continuous case of S-PRESSO vs. continuous audio compression baselines at equivalent compression rates

R

	Method	kbps	Framerate	$M$	FAD $\downarrow$	FAD ${}_{\text{CLAP}}$ $\downarrow$	KAD ${}_{\text{CLAP}}$ $\downarrow$	CLAP ${}_{\text{audio}}$ $\uparrow$	Si-SDR $\uparrow$
Low bitrates	DAC	1.7	86 Hz	2	3.24	0.108	1.71	0.63	-4.11
	SemantiCodec	1.4	100 Hz	1	1.79	0.136	4.93	0.60	-31.8
	S-PRESSO	1.32	11 Hz	12	0.55	0.048	0.728	0.73	-4.48
Ultra low bitrates	SemantiCodec	0.3125	25 Hz	1	1.23	0.271	2.70	0.48	-34.5
	S-PRESSO	0.3	1 Hz	25	0.64	0.052	0.78	0.71	-27.8
	S-PRESSO	0.096	1 Hz	8	0.68	0.060	0.89	0.67	-30.4

Table 2: Performance in the discrete case of S-PRESSO vs. baseline audio codecs at equivalent bitrates.

Figure 3: MUSHRA scores for S-PRESSO, SemantiCodec and a 3.5kHz low-pass anchor, evaluated at $\sim$ 1.35 kbps and $\sim$ 0.3 kbps.

Figure 4: Evaluation of S-PRESSO at varying bitrates and framerates.

Neural Audio Codecs

Neural audio codecs learn compact representations for compressing and reconstructing speech, music, and other sounds.

See all 87 papers

A Generative-First Neural Audio Autoencoder
Feb 17, 2026Jonah Casebeer, Ge Zhu, Zhepei Wang +1Autoencoders Audio Representation Learning
Adobe Research
ZipCodec: Ultra-Low-Frame-Rate Streaming Speech Coding
Sep 10, 2026Luca Della Libera, Cem Subakan, Mirco RavanelliAudio Token Compression Neural Audio Codecs
Concordia University · Mila-Quebec AI Institute · Universit´e Laval
Benchmarking Language Modeling for Lossless Compression of Full-Fidelity Audio
Mar 9, 2026Phillip Long, Zachary Novack, Chris DonahueLossless Data Compression Byte-Level Language Model
University of California, San Diego, Computer Science and Engineering Department · Carnegie Mellon University, School of Computer Science

Feb 17, 2026J/K move · Enter open · S save

A Generative-First Neural Audio Autoencoder

Jonah Casebeer, Ge Zhu, Zhepei Wang, Nicholas J. Bryan

Adobe Research

Autoencoders Audio Representation Learning Neural Audio Codecs Audio Token Compression

Open paper AskPDF arXiv

Neural autoencoders underpin generative models. Practical, large-scale use of neural autoencoders for generative modeling necessitates fast encoding, low latent rates, and a single model across representations. Existing approaches are reconstruction-first: they incur high latent rates, slow encoding, and separate architectures for discrete vs. continuous latents and for different audio channel formats, hindering workflows from preprocessing to inference conditioning. We introduce a generative-first architecture for audio autoencoding that increases temporal downsampling from 2048x to 3360x and supports continuous and discrete representations and common audio channel formats in one model. By balancing compression, quality, and speed, it delivers 10x faster encoding, 1.6x lower rates, and eliminates channel-format-specific variants while maintaining competitive reconstruction quality. This enables applications previously constrained by processing costs: a 60-second mono signal compresses to 788 tokens, making generative modeling more tractable.

Abstract

Neural Audio Codecs

Explore similar work

A Generative-First Neural Audio Autoencoder

ZipCodec: Ultra-Low-Frame-Rate Streaming Speech Coding

Benchmarking Language Modeling for Lossless Compression of Full-Fidelity Audio