mlsanityData to 14 Oct 2026

BitTTS: Highly Compact Text-to-Speech Using 1.58-bit Quantization and Weight Indexing

arXiv 2506.03515 ↗First posted 4 Jun 2025eess.AS cs.LG cs.SDcited by 2

Not on the map. Why?

Abstract (arXiv)

This paper proposes a highly compact, lightweight text-to-speech (TTS) model for on-device applications. To reduce the model size, the proposed model introduces two techniques. First, we introduce quantization-aware training (QAT), which quantizes model parameters during training to as low as 1.58-bit. In this case, most of 32-bit model parameters are quantized to ternary values {-1, 0, 1}. Second, we propose a method named weight indexing. In this method, we save a group of 1.58-bit weights as a single int8 index. This allows for efficient storage of model parameters, even on hardware that treats values in units of 8-bit. Experimental results demonstrate that the proposed method achieved 83 % reduction in model size, while outperforming the baseline of similar model size without quantization in synthesis quality.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 31 of its 39 references that are in the map. First posted on arXiv 4 Jun 2025.

Builds on

31 of 39 references.

References · 31

  1. 2017Attention Is All You Needcited by 60,361
  2. 2016Layer Normalizationcited by 5,669
  3. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  4. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  5. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
  6. 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,101
  7. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 849
  8. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 815
  9. 2021A White Paper on Neural Network Quantizationcited by 472
  10. 2024The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bitscited by 308
  11. 2020Binary Neural Networks: A Surveycited by 190
  12. 2023BitNet: Scaling 1-bit Transformers for Large Language Modelscited by 189
19 more
  1. 2021A Survey on Neural Speech Synthesiscited by 184
  2. 2018Neural Architecture Optimizationcited by 167
  3. 2019An Exponential Learning Rate Schedule for Deep Learningcited by 121
  4. 2023LibriTTS-R: A Restored Multi-Speaker Text-to-Speech Corpuscited by 101
  5. 2019XNOR-Net++: Improved Binary Neural Networkscited by 97
  6. 2018ClariNet: Parallel Wave Generation in End-to-End Text-to-Speechcited by 70
  7. 2019ESPnet-TTS: Unified, Reproducible, and Integratable Open Source End-to-End Text-to-Speech Toolkitcited by 60
  8. 2023A Review of Deep Learning Techniques for Speech Processingcited by 52
  9. 2022An Overview of Affective Speech Synthesis and Conversion in the Deep Learning Eracited by 33
  10. 2021LightSpeech: Lightweight and Fast Text to Speech with Neural Architecture Searchcited by 28
  11. 2022JETS: Jointly Training FastSpeech2 and HiFi-GAN for End to End Text to Speechcited by 26
  12. 20224-bit Conformer with Native Quantization Aware Training for Speech Recognitioncited by 24
  13. 2022Basic Binary Convolution Unit for Binarized Image Restoration Networkcited by 22
  14. 2020SpeedySpeech: Efficient Neural Speech Synthesiscited by 15
  15. 2021Phone-Level Prosody Modelling with GMM-Based MDN for Diverse and Controllable Speech Synthesiscited by 10
  16. 20232-bit Conformer quantization for automatic speech recognitioncited by 10
  17. 2022Lightweight and High-Fidelity End-to-End Text-to-Speech with Multi-Band Generation and Inverse Short-Time Fourier Transformcited by 9
  18. 2022Sub-8-Bit Quantization Aware Training for 8-Bit Neural Network Accelerator with On-Device Speech Recognitioncited by 5
  19. 2022Nix-TTS: Lightweight and End-to-End Text-to-Speech via Module-wise Distillationcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Evolution Strategy-Based Calibration for Low-Bit Quantization of Speech Modelscited by 0
  2. 2026Same Bit Width, Different Outcomes: Post-Training Quantization of Text-to-Speech Across Architecturescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024MobileQuant: Mobile-friendly Quantization for On-device Language Models
  2. 2024Optimizing Large Language Models through Quantization: A Comparative Analysis of PTQ and QAT Techniques
  3. 2024Squat: Quant Small Language Models on the Edge
  4. 2022Sub-8-bit quantization for on-device speech recognition: a regularization-free approach
  5. 2025Edge-ASR: Towards Low-Bit Quantization of Automatic Speech Recognition Models
  6. 2022Efficient Speech Representation Learning with Low-Bit Quantization
  7. 2026ScaleBITS: Scalable Bitwidth Search for Hardware-Aligned Mixed-Precision LLMs
  8. 2025IndexTTS: An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System