A Language Modeling Approach to Diacritic-Free Hebrew TTS
Not on the map. Why?
Abstract (arXiv)
We tackle the task of text-to-speech (TTS) in Hebrew. Traditional Hebrew contains Diacritics, which dictate the way individuals should pronounce given words, however, modern Hebrew rarely uses them. The lack of diacritics in modern Hebrew results in readers expected to conclude the correct pronunciation and understand which phonemes to use based on the context. This imposes a fundamental challenge on TTS systems to accurately map between text-to-speech. In this work, we propose to adopt a language modeling Diacritics-Free approach, for the task of Hebrew TTS. The model operates on discrete speech representations and is conditioned on a word-piece tokenizer. We optimize the proposed method using in-the-wild weakly supervised data and compare it to several diacritic-based TTS systems. Results suggest the proposed method is superior to the evaluated baselines considering both content preservation and naturalness of the generated speech. Samples can be found under the following link: pages.cs.huji.ac.il/adiyoss-lab/HebTTS/
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 27 of its 35 references that are in the map. First posted on arXiv 16 Jul 2024.
Builds on
27 of 35 references.
References · 27
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 4,057
- 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
- 2021WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processingcited by 1,657
- 2016Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translationcited by 1,475
- 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
- 2021SoundStream: An End-to-End Neural Audio Codeccited by 771
- 2023Neural Codec Language Models are Zero-Shot Text to Speech Synthesizerscited by 752
- 2022High Fidelity Neural Audio Compressioncited by 687
- 2022AudioLM: a Language Modeling Approach to Audio Generationcited by 616
- 2024Large Language Models: A Surveycited by 525
- 2023Simple and Controllable Music Generationcited by 488
15 more
- 2023Scaling Speech Technology to 1,000+ Languagescited by 411
- 2022AudioGen: Textually Guided Audio Generationcited by 295
- 2021Speech Resynthesis from Discrete Disentangled Self-Supervised Representationscited by 243
- 2021A Survey on Neural Speech Synthesiscited by 184
- 2023Speak, Read and Prompt: High-Fidelity Text-to-Speech with Minimal Supervisioncited by 169
- 2021Text-Free Prosody-Aware Generative Spoken Language Modelingcited by 119
- 2022I Hear Your True Colors: Image Guided Audio Generationcited by 92
- 2023Zipformer: A faster and better encoder for automatic speech recognitioncited by 80
- 2024Natural language guidance of high-fidelity text-to-speech with synthetic annotationscited by 67
- 2021AlephBERT:A Hebrew Large Pre-Trained Language Model to Start-off your Hebrew NLP Application Withcited by 28
- 2021fairseq S^2: A Scalable and Integrable Speech Synthesis Toolkitcited by 24
- 2023From Discrete Tokens to High-Fidelity Audio Using Multi-Band Diffusioncited by 22
- 2022OverFlow: Putting flows on top of neural transducers for better TTScited by 8
- 2023ivrit.ai: A Comprehensive Dataset of Hebrew Speech for AI Research and Developmentcited by 4
- 2024HebDB: a Weakly Supervised Dataset for Hebrew Speech Processingcited by 3
Built on by
Citing papers per year: 2025: 2.
2 papers.