VibeVoice-ASR-BitNet Technical Report

• URL: http://arxiv.org/abs/2607.21075v2
• Date: Sat, 25 Jul 2026 09:14:28 GMT
• Title: VibeVoice-ASR-BitNet Technical Report
• Abstract summary: VibeVoice-ASR-BitNet is a compressed variant of VibeVoice-ASR optimized for real-time inference on edge CPUs.<n>We implement a progressive quantization-aware training strategy to preserve accuracy under aggressive compression.<n> VibeVoice-ASR-BitNet is 1.6-2.3x faster than Whisper at comparable model sizes.
• Score: 86.93881874028209
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present VibeVoice-ASR-BitNet, a compressed variant of VibeVoice-ASR optimized for real-time inference on edge CPUs. We apply heterogeneous quantization tailored to the computational characteristics of each stage: the VAE acoustic tokenizer uses full-pipeline INT8 quantization (I8_S) with kernel fusion and SIMD optimization, while the autoregressive language model adopts BitNet-style ternary weights (I2_S). To preserve accuracy under aggressive compression, we employ a progressive quantization-aware training strategy. For inference, we implement custom SIMD kernels and fused operators within the ggml framework targeting both ARM and x86 platforms, achieving real-time recognition (RTF < 1) on low-thread-count CPUs. VibeVoice-ASR-BitNet is 1.6--2.3x faster than Whisper.cpp at comparable model sizes (~1.6 GB), with only modest accuracy degradation compared to the FP16 baseline.

Related papers

• Pushing the Limits of On-Device Streaming ASR: A Compact, High-Accuracy English Model for Low-Latency Inference [5.685908474026849]
  High-quality automatic speech recognition (ASR) on edge devices requires models that jointly optimize accuracy, latency, and memory footprint while operating entirely on CPU without GPU acceleration.<n>We conduct a systematic empirical study of state-of-the-art ASR architectures, encompassing encoder-decoder, transducer, and LLM-based paradigms, evaluated across batch, chunked, and streaming inference modes.<n>Our recommended configuration, the int4 k-quant variant, achieves 8.20% average streaming WER across eight standard benchmarks, running comfortably faster than real-time on CPU with 0.56 s algorithmic latency.
  arXiv  Detail & Related papers  (2026-04-16T00:04:32Z)
• Sparse-BitNet: 1.58-bit LLMs are Naturally Friendly to Semi-Structured Sparsity [100.07626315557599]
  We show that 1.58-bit BitNet is naturally more compatible with N:M sparsity than full-precision models.<n>We propose Sparse-BitNet, a unified framework that jointly applies 1.58-bit quantization and dynamic N:M sparsification.
  arXiv  Detail & Related papers  (2026-03-05T13:37:50Z)
• Binary Quantization For LLMs Through Dynamic Grouping [13.578307208515819]
  Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of Natural Language Processing (NLP) tasks.<n> Binary quantization, which compresses model weights from 16-bit Brain Float to 1-bit representations in -1, 1, offers significant reductions in storage and inference costs.<n>We propose a novel optimization objective tailored for binary quantization, along with three algorithms designed to realize it effectively.
  arXiv  Detail & Related papers  (2025-09-03T06:36:21Z)
• Binary and Ternary Natural Language Generation [24.295815261826153]
  Ternary and binary neural networks enable multiplication-free computation. They promise multiple orders of magnitude efficiency gains over full-precision networks. However, such networks have proven very difficult to optimize. We show first ternary and binary transformer models on the downstream tasks of summarization and machine translation.
  arXiv  Detail & Related papers  (2023-06-02T18:01:02Z)
• BiFSMNv2: Pushing Binary Neural Networks for Keyword Spotting to Real-Network Performance [54.214426436283134]
  Deep neural networks, such as the Deep-FSMN, have been widely studied for keyword spotting (KWS) applications. We present a strong yet efficient binary neural network for KWS, namely BiFSMNv2, pushing it to the real-network accuracy performance. We highlight that benefiting from the compact architecture and optimized hardware kernel, BiFSMNv2 can achieve an impressive 25.1x speedup and 20.2x storage-saving on edge hardware.
  arXiv  Detail & Related papers  (2022-11-13T18:31:45Z)
• An Adaptive Device-Edge Co-Inference Framework Based on Soft Actor-Critic [72.35307086274912]
  High-dimension parameter model and large-scale mathematical calculation restrict execution efficiency, especially for Internet of Things (IoT) devices. We propose a new Deep Reinforcement Learning (DRL)-Soft Actor Critic for discrete (SAC-d), which generates the emphexit point, emphexit point, and emphcompressing bits by soft policy iterations. Based on the latency and accuracy aware reward design, such an computation can well adapt to the complex environment like dynamic wireless channel and arbitrary processing, and is capable of supporting the 5G URL
  arXiv  Detail & Related papers  (2022-01-09T09:31:50Z)
• 8-bit Optimizers via Block-wise Quantization [57.25800395197516]
  Statefuls maintain statistics over time, e.g., the exponentially smoothed sum (SGD with momentum) or squared sum (Adam) of past values. This state can be used to accelerate optimization compared to plain gradient descent but uses memory that might otherwise be allocated to model parameters. In this paper, we develop first gradients that use 8-bit statistics while maintaining the performance levels of using 32-bit gradient states.
  arXiv  Detail & Related papers  (2021-10-06T15:43:20Z)
• DS-Net++: Dynamic Weight Slicing for Efficient Inference in CNNs and Transformers [105.74546828182834]
  We show a hardware-efficient dynamic inference regime, named dynamic weight slicing, which adaptively slice a part of network parameters for inputs with diverse difficulty levels. We present dynamic slimmable network (DS-Net) and dynamic slice-able network (DS-Net++) by input-dependently adjusting filter numbers of CNNs and multiple dimensions in both CNNs and transformers.
  arXiv  Detail & Related papers  (2021-09-21T09:57:21Z)
• Real-Time Execution of Large-scale Language Models on Mobile [49.32610509282623]
  We find the best model structure of BERT for a given computation size to match specific devices. Our framework can guarantee the identified model to meet both resource and real-time specifications of mobile devices. Specifically, our model is 5.2x faster on CPU and 4.1x faster on GPU with 0.5-2% accuracy loss compared with BERT-base.
  arXiv  Detail & Related papers  (2020-09-15T01:59:17Z)
• TASO: Time and Space Optimization for Memory-Constrained DNN Inference [5.023660118588569]
  Convolutional neural networks (CNNs) are used in many embedded applications, from industrial robotics and automation systems to biometric identification on mobile devices. We propose an approach for ahead-of-time domain specific optimization of CNN models, based on an integer linear programming (ILP) for selecting primitive operations to implement convolutional layers.
  arXiv  Detail & Related papers  (2020-05-21T15:08:06Z)
• ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context [58.40112382877868]
  We propose a novel CNN-RNN-transducer architecture, which we call ContextNet. ContextNet features a fully convolutional encoder that incorporates global context information into convolution layers by adding squeeze-and-excitation modules. We demonstrate that ContextNet achieves a word error rate (WER) of 2.1%/4.6% without external language model (LM), 1.9%/4.1% with LM and 2.9%/7.0% with only 10M parameters on the clean/noisy LibriSpeech test sets.
  arXiv  Detail & Related papers  (2020-05-07T01:03:18Z)
• FastWave: Accelerating Autoregressive Convolutional Neural Networks on FPGA [27.50143717931293]
  WaveNet is a deep autoregressive CNN composed of several stacked layers of dilated convolution. We develop the first accelerator platformtextitFastWave for autoregressive convolutional neural networks.
  arXiv  Detail & Related papers  (2020-02-09T06:15:09Z)

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.