REVIEW 8 cited by
Scaling Transformers for Low-Bitrate High-Quality Speech Coding
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
abstract
The tokenization of speech with neural audio codec models is a vital part of modern AI pipelines for the generation or understanding of speech, alone or in a multimodal context. Traditionally such tokenization models have concentrated on low parameter-count architectures using only components with strong inductive biases. In this work we show that by scaling a transformer architecture with large parameter count to this problem, and applying a flexible Finite Scalar Quantization (FSQ) based bottleneck, it is possible to reach state-of-the-art speech quality at extremely low bit-rates of $400$ or $700$ bits-per-second. The trained models strongly out-perform existing baselines in both objective and subjective tests.
Forward citations
Cited by 8 Pith papers
-
Next Tokens Denoising for Speech Synthesis
Dragon-FM generates speech autoregressively over two-second chunks while using flow matching inside each chunk, achieving fast synthesis at 12.5 discrete audio tokens per second.
-
Probing the Robustness Properties of Neural Speech Codecs
DAC is the most noise-robust neural codec at high bitrates, but at 3 kbps EnCodec wins, and measured non-linearity correlates with robustness.
-
CoDiCodec: Unifying Continuous and Discrete Compressed Representations of Audio
CoDiCodec unifies continuous and discrete audio compression in one consistency-trained autoencoder, using FSQ-dropout to serve both continuous ~11 Hz embeddings and 2.38 kbps discrete tokens.
-
CodecBench: A Comprehensive Benchmark for Acoustic and Semantic Evaluation
CodecBench ranks 14 audio codecs on acoustic fidelity and semantic preservation across 19 datasets and four audio domains, revealing a reconstruction-versus-semantics tradeoff.
-
NanoCodec: Towards High-Quality Ultra Fast Speech LLM Inference
NanoCodec achieves competitive speech quality at 12.5 frames per second and 0.6-1.78 kbps, with a causal decoder for low-latency speech LLM inference.
-
Quantize More, Lose Less: Autoregressive Generation from Residually Quantized Speech Representations
QTTS models speech as sequences from a multi-codebook RVQ audio codec whose first codebook is trained with ASR supervision, aiming for higher-fidelity TTS than single-codebook systems.
-
UniTTS: An end-to-end TTS system without decoupling of acoustic and semantic information
The authors propose DistilCodec, a 32,768-code single-codebook audio codec, and UniTTS, a Qwen2.5-7B TTS model trained with audio, text, and cross-modal autoregressive tasks on interleaved prompts.
-
Robust Residual Finite Scalar Quantization for Neural Compression
RFSQ applies learned scaling or invertible LayerNorm to residual FSQ to prevent magnitude decay, reporting DNSMOS and image loss gains, though the LayerNorm variant has a reconstruction inconsistency.
Discussion (0). Continue with ORCID to comment.