Papers
arxiv:2608.05727

LILAC: An Idempotent Neural Speech Codec

Published on Aug 6
Authors:
,
,
,

Abstract

LILAC is a fully convolutional neural audio codec that guarantees idempotency by construction while maintaining competitive speech quality at very low bitrates.

Neural Audio Codecs are widely adopted in speech generation and editing. However, existing neural audio codecs are not idempotent: across the paper's twelve baseline systems, every configuration tested rewrites, on average, at least 15% of its tokens in a single decode-re-encode pass. This poses a problem for utilizing Neural Audio Codecs as token interfaces in pipelines where re-encoding decoded outputs can occur. We present LILAC, a fully convolutional 24 kHz speech codec at 9.375 Hz and 0.75 kbit/s that is codec idempotent by construction; re-encoding the decoded audio of any valid token stream returns the identical stream. LILAC achieves idempotency while maintaining competitive quality, reaching UTMOS 4.14 and 4.24 on LibriSpeech and LibriTTS-R test sets, comparable to SOTA sub-1 kbit/s Neural Audio Codecs.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2608.05727
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 1

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2608.05727 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2608.05727 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.