Stable Autoregressive Speech Generation with Low-Frame-Rate High-Dimensional Continuous Tokens

• URL: http://arxiv.org/abs/2607.29363v1
• Date: Fri, 31 Jul 2026 12:51:24 GMT
• Title: Stable Autoregressive Speech Generation with Low-Frame-Rate High-Dimensional Continuous Tokens
• Abstract summary: We show how a low-frame-rate, high-dimensional, high-bandwidth continuous representation can be co-designed with a streaming generation framework.<n>Our design preserves reconstruction quality, improves single-token predictability, achieves competitive WER, and maintains stable long-form synthesis.
• Score: 29.85404709913323
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Balancing sequence length, representational capacity, and long-horizon stability is a central problem in autoregressive (AR) speech and audio generation. Representations with higher frame rates or greater capacity can preserve more signal detail, but they also make streaming generation more vulnerable to distribution drift and AR error accumulation. Conversely, shorter and more compressed representations simplify AR modeling, but their limited bandwidth may discard important components and constrain the upper bound of reconstruction fidelity and generation quality. We ask whether a low-frame-rate, high-dimensional, high-bandwidth continuous representation can be co-designed with a streaming generation framework to support robust high-fidelity reconstruction, strong single-token predictability, and superior long-horizon stability. We decompose this goal into two coupled problems: what geometric and statistical properties a high-dimensional representation space should have, and how an AR continuous-token generator should be structured to resist error accumulation. Accordingly, we propose Locodec, a locally encoded codec that shapes its representation space to improve the interpolatability of a lower-dimensional core manifold and the identifiability of the native high-dimensional coordinates, thereby improving the predictability of high-dimensional high-bandwidth tokens. We also propose MP-ELD, a single-token AR flow-matching framework that uses multi-path information routing and residual classifier-free guidance to mitigate error accumulation. Experiments with 8-Hz, 768-dimensional tokens show that our design preserves reconstruction quality, improves single-token predictability, achieves competitive WER, and maintains stable long-form synthesis, without using external SSL/ASR models, pretrained text language models, or post-training stages.

Related papers

• Rarity-Aware Discrete Diffusion with Spatially Consistent Decoding for Photo-Realistic Image Super-Resolution [80.28905986756685]
  DiMOO-SR is a rarity-aware multimodal discrete diffusion framework for photo-realistic image Super-Resolution.<n>Experiments on widely used real-world SR benchmarks demonstrate that DiMOO-SR achieves competitive perceptual quality with only a few parallel decoding steps.
  arXiv  Detail & Related papers  (2026-07-20T07:01:01Z)
• Where to Refine, When to Stop: Rethinking Redundancy via Latent Discrepancy for Efficient Visual Autoregressive Generation [50.139984798361375]
  We propose a training-free framework that removes redundancy via latent discrepancy.<n> LD-Pruning substantially reduces inference while maintaining high generation quality, achieving up to 2.35x speedup on Infinity-8B.
  arXiv  Detail & Related papers  (2026-05-29T19:34:39Z)
• Improving Reconstruction of Representation Autoencoder [52.817427902597416]
  We propose LV-RAE, a representation autoencoder that augments semantic features with missing low-level information.<n>Our experiments demonstrate that LV-RAE significantly improves reconstruction fidelity while preserving the semantic abstraction.
  arXiv  Detail & Related papers  (2026-02-09T13:12:35Z)
• ResTok: Learning Hierarchical Residuals in 1D Visual Tokenizers for Autoregressive Image Generation [64.84095852784714]
  Residual Tokenizer (ResTok) is a 1D visual tokenizer that builds hierarchical residuals for both image tokens and latent tokens.<n>We show that restoring hierarchical residual priors in visual tokenization significantly improves AR image generation, achieving a gFID of 2.34 on ImageNet-256 with only 9 sampling steps.
  arXiv  Detail & Related papers  (2026-01-07T14:09:18Z)
• LUMA-RAG: Lifelong Multimodal Agents with Provably Stable Streaming Alignment [0.0]
  Retrieval-Augmented Generation has emerged as the dominant paradigm for grounding large language model outputs in verifiable evidence.<n>We present LUMA-RAG, a lifelong multimodal agent architecture featuring three key innovations.<n> Experiments demonstrate robust text-to-image retrieval (Recall@10 = 0.94), graceful performance degradation under product quantization offloading, and provably stable audio-to-image rankings (Safe@1 = 1.0)
  arXiv  Detail & Related papers  (2025-11-04T08:47:12Z)
• Hawk: Leveraging Spatial Context for Faster Autoregressive Text-to-Image Generation [87.00172597953228]
  Speculative decoding has shown promise in accelerating text generation without compromising quality.<n>We introduce Hawk, a new approach that harnesses the spatial structure of images to guide the speculative model toward more accurate and efficient predictions.<n> Experimental results on multiple text-to-image benchmarks demonstrate a 1.71x speedup over standard AR models.
  arXiv  Detail & Related papers  (2025-10-29T17:43:31Z)
• Next Tokens Denoising for Speech Synthesis [51.320443764269726]
  Dragon-FM is a novel text-to-speech (TTS) design that unifies AR and flow-matching.<n>It processes 48 kHz audio tokens in chunks at a compact rate of 12.5 tokens per second.<n>Experiments on podcast datasets demonstrate its capability to efficiently generate high-quality zero-shot podcasts.
  arXiv  Detail & Related papers  (2025-07-30T15:03:36Z)
• Frequency-Aware Autoregressive Modeling for Efficient High-Resolution Image Synthesis [40.93077975823353]
  Visual autoregressive modeling, based on the next-scale prediction paradigm, exhibits notable advantages in image quality and model scalability.<n>However, the computational overhead in high-resolution stages remains a critical challenge due to the substantial number of tokens involved.<n>We introduce Sparsevar, a plug-and-play acceleration framework for next-scale prediction that dynamically excludes low-frequency tokens during inference without requiring additional training.
  arXiv  Detail & Related papers  (2025-07-28T01:13:24Z)
• Quantize More, Lose Less: Autoregressive Generation from Residually Quantized Speech Representations [26.938560887095658]
  Existing autoregressive approaches often rely on single-codebook representations, which suffer from significant information loss.<n>We propose QTTS, a novel TTS framework built upon our new audio, QDAC.<n>Our experiments demonstrate that the proposed framework achieves higher synthesis quality and better preserves expressive content compared to baseline.
  arXiv  Detail & Related papers  (2025-07-16T12:47:09Z)
• Fast Autoregressive Models for Continuous Latent Generation [49.079819389916764]
  Autoregressive models have demonstrated remarkable success in sequential data generation, particularly in NLP.<n>Recent work, the masked autoregressive model (MAR) bypasses quantization by modeling per-token distributions in continuous spaces using a diffusion head.<n>We propose Fast AutoRegressive model (FAR), a novel framework that replaces MAR's diffusion head with a lightweight shortcut head.
  arXiv  Detail & Related papers  (2025-04-24T13:57:08Z)
• CD-NGP: A Fast Scalable Continual Representation for Dynamic Scenes [31.783117836434403]
  CD-NGP is a continual learning framework that reduces memory overhead and enhances scalability.<n>It significantly reduces training memory usage to 14GB and requires only 0.4MB/frame in streaming bandwidth on DyNeRF.
  arXiv  Detail & Related papers  (2024-09-08T17:35:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.