# LightSpeech: Lightweight and Fast Text to Speech with Neural Architecture Search

arXiv: https://arxiv.org/abs/2102.04040 · HTML: https://mlsanity.com/paper/2102.04040/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 24 references that are in the map.
- First posted on arXiv 8 Feb 2021.

## Builds on (most cited first, 17 of 17)

- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Learning Transferable Architectures for Scalable Image Recognition](https://mlsanity.com/paper/1707.07012/) (2017)
- [Efficient Neural Architecture Search via Parameter Sharing](https://mlsanity.com/paper/1802.03268/) (2018)
- [Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions](https://mlsanity.com/paper/1712.05884/) (2017)
- [FastSpeech 2: Fast and High-Quality End-to-End Text to Speech](https://mlsanity.com/paper/2006.04558/) (2020)
- [Channel Pruning for Accelerating Very Deep Neural Networks](https://mlsanity.com/paper/1707.06168/) (2017)
- [BinaryConnect: Training Deep Neural Networks with binary weights during propagations](https://mlsanity.com/paper/1511.00363/) (2015)
- [Tacotron: Towards End-to-End Speech Synthesis](https://mlsanity.com/paper/1703.10135/) (2017)
- [Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram](https://mlsanity.com/paper/1910.11480/) (2019)
- [Tensor Decomposition for Signal Processing and Machine Learning](https://mlsanity.com/paper/1607.01668/) (2016)
- [Neural Architecture Optimization](https://mlsanity.com/paper/1808.07233/) (2018)
- [Neural Predictor for Neural Architecture Search](https://mlsanity.com/paper/1912.00848/) (2019)
- [AdaBERT: Task-Adaptive BERT Compression with Differentiable Neural Architecture Search](https://mlsanity.com/paper/2001.04246/) (2020)
- [Semi-Supervised Neural Architecture Search](https://mlsanity.com/paper/2002.10389/) (2020)
- [Depthwise Separable Convolutions for Neural Machine Translation](https://mlsanity.com/paper/1706.03059/) (2017)
- [JDI-T: Jointly trained Duration Informed Transformer for Text-To-Speech without Explicit Alignment](https://mlsanity.com/paper/2005.07799/) (2020)
- [Non-Autoregressive Neural Text-to-Speech](https://mlsanity.com/paper/1905.08459/) (2019)

## Built on by (most cited first, 28 of 28)

- [A Survey on Neural Speech Synthesis](https://mlsanity.com/paper/2106.15561/) (2021)
- [AdaSpeech: Adaptive Text to Speech for Custom Voice](https://mlsanity.com/paper/2103.00993/) (2021)
- [A Survey of Techniques for Optimizing Transformer Inference](https://mlsanity.com/paper/2307.07982/) (2023)
- [ZipVoice: Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching](https://mlsanity.com/paper/2506.13053/) (2025)
- [LightGrad: Lightweight Diffusion Probabilistic Model for Text-to-Speech](https://mlsanity.com/paper/2308.16569/) (2023)
- [Lightweight and High-Fidelity End-to-End Text-to-Speech with Multi-Band Generation and Inverse Short-Time Fourier Transform](https://mlsanity.com/paper/2210.15975/) (2022)
- [Evaluating and reducing the distance between synthetic and real speech distributions](https://mlsanity.com/paper/2211.16049/) (2022)
- [EMNS /Imz/ Corpus: An emotive single-speaker dataset for narrative storytelling in games, television and graphic novels](https://mlsanity.com/paper/2305.13137/) (2023)
- [Neural Architecture Search for Speech Emotion Recognition](https://mlsanity.com/paper/2203.16928/) (2022)
- [SAPIEN: Affective Virtual Agents Powered by Large Language Models](https://mlsanity.com/paper/2308.03022/) (2023)
- [Nix-TTS: Lightweight and End-to-End Text-to-Speech via Module-wise Distillation](https://mlsanity.com/paper/2203.15643/) (2022)
- [EfficientSpeech: An On-Device Text to Speech Model](https://mlsanity.com/paper/2305.13905/) (2023)
- [EfficientTDNN: Efficient Architecture Search for Speaker Recognition](https://mlsanity.com/paper/2103.13581/) (2021)
- [On the Interplay Between Sparsity, Naturalness, Intelligibility, and Prosody in Speech Synthesis](https://mlsanity.com/paper/2110.01147/) (2021)
- [Evolutionary Neural Architecture Search for Transformer in Knowledge Tracing](https://mlsanity.com/paper/2310.01180/) (2023)
- [Prosody Analysis of Audiobooks](https://mlsanity.com/paper/2310.06930/) (2023)
- [Custom Data Augmentation for low resource ASR using Bark and Retrieval-Based Voice Conversion](https://mlsanity.com/paper/2311.14836/) (2023)
- [SwiftF0: Fast and Accurate Monophonic Pitch Detection](https://mlsanity.com/paper/2508.18440/) (2025)
- [Designing Novel Cognitive Diagnosis Models via Evolutionary Multi-Objective Neural Architecture Search](https://mlsanity.com/paper/2307.04429/) (2023)
- [BitTTS: Highly Compact Text-to-Speech Using 1.58-bit Quantization and Weight Indexing](https://mlsanity.com/paper/2506.03515/) (2025)
- [HyperTTS: Parameter Efficient Adaptation in Text to Speech using Hypernetworks](https://mlsanity.com/paper/2404.04645/) (2024)
- [Lightweight End-to-end Text-to-speech Synthesis for low resource on-device applications](https://mlsanity.com/paper/2505.07701/) (2025)
- [Cross-lingual Knowledge Distillation via Flow-based Voice Conversion for Robust Polyglot Text-To-Speech](https://mlsanity.com/paper/2309.08255/) (2023)
- [CrossSpeech++: Cross-lingual Speech Synthesis with Decoupled Language and Speaker Generation](https://mlsanity.com/paper/2412.20048/) (2024)
- [Unseen Speaker and Language Adaptation for Lightweight Text-To-Speech with Adapters](https://mlsanity.com/paper/2508.18006/) (2025)
- [Performance and Complexity Trade-off Optimization of Speech Models During Training](https://mlsanity.com/paper/2601.13704/) (2026)
- [Same Bit Width, Different Outcomes: Post-Training Quantization of Text-to-Speech Across Architectures](https://mlsanity.com/paper/2609.28974/) (2026)
- [Paradee: Distilling Kokoro-82M into an 8M-Parameter Single-Voice Text-to-Speech Model](https://mlsanity.com/paper/2610.06817/) (2026)

## Abstract (arXiv)

Text to speech (TTS) has been broadly used to synthesize natural and intelligible speech in different scenarios. Deploying TTS in various end devices such as mobile phones or embedded devices requires extremely small memory usage and inference latency. While non-autoregressive TTS models such as FastSpeech have achieved significantly faster inference speed than autoregressive models, their model size and inference latency are still large for the deployment in resource constrained devices. In this paper, we propose LightSpeech, which leverages neural architecture search~(NAS) to automatically design more lightweight and efficient models based on FastSpeech. We first profile the components of current FastSpeech model and carefully design a novel search space containing various lightweight and potentially effective architectures. Then NAS is utilized to automatically discover well performing architectures within the search space. Experiments show that the model discovered by our method achieves 15x model compression ratio and 6.5x inference speedup on CPU with on par voice quality. Audio demos are provided at https://speechresearch.github.io/lightspeech.