mlsanityData to 7 Oct 2026

VITS2: Improving Quality and Efficiency of Single-Stage Text-to-Speech with Adversarial Learning and Architecture Design

arXiv 2307.16430 ↗First posted 31 Jul 2023cs.SD cs.LG eess.AScited by 25

Not on the map. Why?

Abstract (arXiv)

Single-stage text-to-speech models have been actively studied recently, and their results have outperformed two-stage pipeline systems. Although the previous single-stage model has made great progress, there is room for improvement in terms of its intermittent unnaturalness, computational efficiency, and strong dependence on phoneme conversion. In this work, we introduce VITS2, a single-stage text-to-speech model that efficiently synthesizes a more natural speech by improving several aspects of the previous work. We propose improved structures and training mechanisms and present that the proposed methods are effective in improving naturalness, similarity of speech characteristics in a multi-speaker model, and efficiency of training and inference. Furthermore, we demonstrate that the strong dependence on phoneme conversion in previous works can be significantly reduced with our method, which allows a fully end-to-end single-stage approach.

Key facts. Cited by 25 papers in the map (data through 7 Oct 2026). Builds on 19 of its 24 references that are in the map. First posted on arXiv 31 Jul 2023.

Builds on

19 of 24 references.

References · 19

  1. 2016WaveNet: A Generative Model for Raw Audiocited by 1,710
  2. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,139
  3. 2016Least Squares Generative Adversarial Networkscited by 1,000
  4. 2017Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictionscited by 800
  5. 2020FastSpeech 2: Fast and High-Quality End-to-End Text to Speechcited by 710
  6. 2020WaveGrad: Estimating Gradients for Waveform Generationcited by 620
  7. 2021Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speechcited by 555
  8. 2021Grad-TTS: A Diffusion Probabilistic Model for Text-to-Speechcited by 445
  9. 2019MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesiscited by 351
  10. 2019Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogramcited by 337
  11. 2018WaveGlow: A Flow-based Generative Network for Speech Synthesiscited by 308
  12. 2018Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesiscited by 280
7 more
  1. 2020Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Searchcited by 269
  2. 2018Efficient Neural Audio Synthesiscited by 218
  3. 2022NaturalSpeech: End-to-End Text to Speech Synthesis with Human-Level Qualitycited by 151
  4. 2019High Fidelity Speech Synthesis with Adversarial Networkscited by 90
  5. 2020End-to-End Adversarial Text-to-Speechcited by 77
  6. 2020Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesiscited by 54
  7. 2022JETS: Jointly Training FastSpeech2 and HiFi-GAN for End to End Text to Speechcited by 26

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 6; 2025: 15; 2026: 3.

25 papers.

Sort:

By area:

  1. 2024WavChat: A Survey of Spoken Dialogue Modelscited by 88
  2. 2024SpeechCraft: A Fine-grained Expressive Speech Dataset with Natural Language Descriptioncited by 20
  3. 2025WavReward: Spoken Dialogue Models With Generalist Reward Evaluatorscited by 15
  4. 2025TaoAvatar: Real-Time Lifelike Full-Body Talking Avatars for Augmented Reality via 3D Gaussian Splattingcited by 11
  5. 2025Accelerating Autoregressive Speech Synthesis Inference With Speech Speculative Decodingcited by 6
  6. 2025M3-TTS: Multi-modal DiT Alignment & Mel-latent for Zero-shot High-fidelity Speech Synthesiscited by 6
  7. 2024Fake it to make it: Using synthetic data to remedy the data shortage in joint multimodal speech-and-gesture synthesiscited by 5
  8. 2025DiEmo-TTS: Disentangled Emotion Representations via Self-Supervised Distillation for Cross-Speaker Emotion Transfer in Text-to-Speechcited by 5
  9. 2025FaceSpeak: Expressive and High-Quality Speech Synthesis from Human Portraits of Different Stylescited by 4
  10. 2025MathReader : Text-to-Speech for Mathematical Documentscited by 3
  11. 2025FMSD-TTS: Few-shot Multi-Speaker Multi-Dialect Text-to-Speech Synthesis for "U-Tsang, Amdo and Kham Speech Dataset Generationcited by 3
  12. 2024Multi-modal Adversarial Training for Zero-Shot Voice Cloningcited by 2
13 more
  1. 2025Towards Better Disentanglement in Non-Autoregressive Zero-Shot Expressive Voice Conversioncited by 2
  2. 2024VCounselor: A Psychological Intervention Chat Agent Based on a Knowledge-Enhanced Large Language Modelcited by 1
  3. 2025Comparative Evaluation of Expressive Japanese Character Text-to-Speech with VITS and Style-BERT-VITS2cited by 1
  4. 2023Rethinking Internet Communication Through LLMs: How Close Are We?cited by 0
  5. 2024A multi-speaker multi-lingual voice cloning system based on vits2 for limmits 2024 challengecited by 0
  6. 2025Clip-TTS: Contrastive Text-content and Mel-spectrogram, A High-Quality Text-to-Speech Method based on Contextual Semantic Understandingcited by 0
  7. 2025Balalaika: Data-Centric, Prosody-Aware Annotation Pipeline for Russian Speechcited by 0
  8. 2025FNH-TTS: Mixture-of-Experts Duration Modeling for Robust Neural Speech Synthesiscited by 0
  9. 2025TMD-TTS: A Unified Tibetan Multi-Dialect Text-to-Speech Framework for "U-Tsang, Amdo and Kham Speech Dataset Generationcited by 0
  10. 2025Flamed-TTS: Flow Matching Attention-Free Models for Efficient Generating and Dynamic Pacing Zero-shot Text-to-Speechcited by 0
  11. 2026Vagdhenu: A Vrutta (Meter) Aware Shloka-to-Chant (TTS) System for Sanskritcited by 0
  12. 2026Morpho-VITS: Variational Inference with Morphological Modeling for End-to-End Speech Synthesis of a Tonal Bantu Languagecited by 0
  13. 2026BanglaKontho: Closing the Long-Form Gap in Bangla Text-to-Speechcited by 0