Pith. sign in

REVIEW 6 cited by

SemantiCodec: An Ultra Low Bitrate Semantic Audio Codec for General Sound

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.00233 v2 pith:DUEPQEX4 submitted 2024-04-30 cs.SD cs.AIcs.MMeess.ASeess.SP

classification cs.SDcs.AIcs.MMeess.ASeess.SP
keywords audiosemanticodecsemanticsignificantlycodeccodecsencoderlanguage
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Large language models (LLMs) have significantly advanced audio processing through audio codecs that convert audio into discrete tokens, enabling the application of language modelling techniques to audio data. However, traditional codecs often operate at high bitrates or within narrow domains such as speech and lack the semantic clues required for efficient language modelling. Addressing these challenges, we introduce SemantiCodec, a novel codec designed to compress audio into fewer than a hundred tokens per second across diverse audio types, including speech, general sound, and music, without compromising quality. SemantiCodec features a dual-encoder architecture: a semantic encoder using a self-supervised pre-trained Audio Masked Autoencoder (AudioMAE), discretized using k-means clustering on extensive audio data, and an acoustic encoder to capture the remaining details. The semantic and acoustic encoder outputs are used to reconstruct audio via a diffusion-model-based decoder. SemantiCodec is presented in three variants with token rates of 25, 50, and 100 per second, supporting a range of ultra-low bit rates between 0.31 kbps and 1.40 kbps. Experimental results demonstrate that SemantiCodec significantly outperforms the state-of-the-art Descript codec on reconstruction quality. Our results also suggest that SemantiCodec contains significantly richer semantic information than all evaluated state-of-the-art audio codecs, even at significantly lower bitrates. Our code and demos are available at https://haoheliu.github.io/SemantiCodec/.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Autoregressive Speech Enhancement via Acoustic Tokens

    cs.SD 2025-07 conditional novelty 6.0 of 10

    Acoustic tokens outperform semantic tokens on speaker identity in speech enhancement, an autoregressive transducer helps in some settings, but discrete representations still lag continuous ones.

  2. StarVC: A Unified Auto-Regressive Framework for Joint Text and Speech Generation in Voice Conversion

    cs.MM 2025-06 conditional novelty 6.0 of 10

    StarVC is an autoregressive voice conversion model that generates text tokens before acoustic tokens, improving linguistic fidelity while retaining speaker similarity.

  3. Probing the Robustness Properties of Neural Speech Codecs

    eess.AS 2025-05 conditional novelty 6.0 of 10

    DAC is the most noise-robust neural codec at high bitrates, but at 3 kbps EnCodec wins, and measured non-linearity correlates with robustness.

  4. GenSE: Generative Speech Enhancement via Language Models using Hierarchical Modeling

    eess.AS 2025-02 conditional novelty 6.0 of 10

    GenSE enhances speech by first denoising semantic tokens with a language model and then generating acoustic tokens from a single-quantizer codec, reporting higher DNSMOS, speaker similarity, and lower WER than prior systems.

  5. CLAP-ART: Automated Audio Captioning with Semantic-rich Audio Representation Tokenizer

    eess.AS 2025-06 conditional novelty 5.0 of 10

    CLAP-ART improves automated audio captioning by feeding BART discrete tokens produced from a semantic audio representation (BEATs) rather than from a waveform codec.

  6. Metis: A Foundation Speech Generation Model with Masked Generative Pre-training

    cs.SD 2025-02 conditional novelty 5.0 of 10

    A masked generative model pre-trained on unlabeled speech then fine-tuned per task matches or beats task-specific systems across TTS, voice conversion, speaker extraction, enhancement, and lip-to-speech.

Pith tools