Text-to-Speech

Also known as TTS

Text-to-speech systems turn text into speech that people can understand and use, with control over voice, accent, emotion, and speaking style. Common approaches include autoregressive models over speech-codec tokens, diffusion and flow-matching models, neural vocoders, and fine-tuning or preference-based post-training. Evaluation combines listening tests with measures such as word error rate and speaker similarity, using benchmarks for zero-shot cloning, multilingual synthesis, and instruction following. A central challenge is balancing intelligibility, speaker identity, and expressive appropriateness: high naturalness scores can miss pronunciation errors or poor fit to a domain, while low-resource languages and fast, robust generation remain difficult.

Latest papers

67 papers4 papers in the last four weeks, down 50% on the four weeks before. 0.0% of all new papers.

SEAM: Spectral Embedding Alignment and Merging for Low-Resource TTS
Oct 19, 2026Walaa Othman, Baher Mohammad, Stamatios Lefkimmiatis +1Text-to-Speech Low-Resource TTS Synthesis
ITMO University · SPC RAS · MWS AI
Does Expert Supervision Improve Text-to-Speech? An Empirical Study of Preference Optimization
Oct 18, 2026Minh Duc Vu, Junchuan Zhao, Ye WangTTS Synthesis Text-to-Speech
Department of Statistics & Data Science, National University of Singapore, Singapore · School of Computing, National University of Singapore, Singapore
Training-Free Instruction TTS Gender Bias Calibration Using Model-Adaptive Steering
Oct 7, 2026Kuan-Yu Chen, Yi-Cheng Lin, Jeng-Lin Li +1Text-to-Speech TTS Synthesis
AI Research Center, Inventec Corporation, Taiwan · Graduate Institute of Communication Engineering, National Taiwan University, Taiwan
BanglaBox: A Phonetically-Balanced Corpus and Data-Efficient Foundation-Model Adaptation for Bangla Text-to-Speech with Zero-Shot Voice Cloning
Oct 6, 2026Emtiaz Uddin Ahmed, Araf Mahmud, Sajib Hossain +3Text-to-Speech Low-Resource TTS Synthesis
Refinement Buys Intelligibility, Search Buys Identity: What Test-Time Compute Buys in Masked-Diffusion TTS
Oct 2, 2026Nityanand Mathur, Hamees Sayed, Ayush Pratap SinghTTS Synthesis Text-to-Speech
Work done outside of Blackstar Inc. · Work done at Smallest AI.
Balalaika-Longform: A Russian Speech Corpus for Continuous Long-Form Text-to-Speech
Sep 30, 2026Nikita Vasiliev, Kirill Borodin, Vasilii Kudryavtsev +2TTS Synthesis TTS Evaluation
BitmanagerAI, Dubai, UAE · lab260, Yerevan, Armenia · MTUCI, Moscow, Russia
RAWD-TTS: Ratio-Free Reward Alignment for Discrete-Diffusion Voice Cloning
Sep 29, 2026Maxim Maslov, Kirill Borodin, Vasilii Kudryavtsev +2Audio Diffusion Models Zero-Shot TTS
lab260, Yerevan, Armenia · BitmanagerAI, Dubai, UAE · MTUCI, Moscow, Russia
DualTrack: Synchronized speech-gesture generation via symmetric coupling of pretrained priors
Sep 29, 2026Yuanzhuo Hu, Zehan Liu, Xiaoyi Qin +1Human Motion Generation Co-Speech Gesture Generation
The Chinese University of Hong Kong, Shenzhen, China · X Square Robot
InstCharVoice: Grounding Natural-Language Instructions for Character-Level Control in Text-to-Speech
Sep 28, 2026Sihang Nie, Xueru Li, Xiaofen Xing +4TTS Synthesis Language Model-Based Control
South China University of Technology · Huya Inc. · The Hongkong Polytechnic University
Accent Analogy Guidance: More Speaker Similarity at Equal Accent in Cross-Lingual Voice Cloning
Sep 24, 2026Yoomee Cho, Jisun LeeTTS Synthesis Multilingual TTS Synthesis
ESTsoft, Seoul, Republic of Korea
Bridging Data, Reasoning, and Alignment: A Unified Framework for Context-Aware Instruction-Following TTS
Sep 13, 2026Jingbin Hu, Luyu Wang, Wenjie Tian +7Text-to-Speech Speech Language Models
Shenzhen Pimei Technology Co., Ltd., Guangdong, China · School of Intelligence Science and Technology, Nanjing University, Suzhou, China · State Key Laboratory of Novel Software Technology, Nanjing University, Nanjing, China
Post-Training Zero-Shot TTS for Fine-Grained Emotion and Duration Control via Natural Language
Sep 10, 2026Lianru Gao, Yujie Guo, Yong QinTTS Synthesis Zero-Shot TTS
TMCC, College of Computer Science, Nankai University, Tianjin, China
Scalable Direction-Following TTS via Voice Impression-Guided Pseudo Triplet Construction
Sep 2, 2026Kenichi Fujita, Yusuke IjimaExpressive Speech Synthesis Text-to-Speech
NTT, Inc., Japan
Diagnose, Then Refine: A Closed-Loop TTS System with AudioLLM-Guided Correction
Aug 29, 2026Zeyang Song, Tianchi Liu, Tianrui Wang +3TTS Synthesis Text-to-Speech
National University of Singapore · Tencent · LIGHTSPEED +4
ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS
Aug 11, 2026Shijun Luo, Lizhi WanASR Evaluation TTS Synthesis
NetEase Cloud Music, Hangzhou, China
Beyond Naturalness: Probing Automated Text-To-Speech Evaluators on Linguistically Grounded Dimensions
Aug 10, 2026Oluwanifemi Bamgbose, Jash Shah, Simon Rosen +6Audio-Language Model Evaluation Human Preference Evaluation
ServiceNow
DialectS2S: End-to-End Speech Dialogue Modeling for Low-Resource Chinese Dialects
Aug 8, 2026Yi Shu, Tianyu Peng, Yingzhuo Deng +5Spoken Dialogue Systems Speech Language Models
School of Artificial Intelligence, University of Chinese Academy of Sciences · Institute of Automation, Chinese Academy of Sciences · Wuhan AI Research +1
Domain-Specific Evaluation of Text-to-Speech Systems: A Multi-Metric Benchmarking Study
Aug 3, 2026Ali Jafar, Amal Sarmad, Shifa Yousaf +1Speech Generation Evaluation TTS Evaluation
FAST School of Computing, National University of Computer and Emerging Sciences (FAST-NUCES), Lahore, Punjab, Pakistan
Faster IndexTTS-2: Accelerating and Streaming Autoregressive Zero-Shot Text-to-Speech Synthesis on GPUs
Jul 23, 2026Muyang Du, Shuang Yu, Junjie LaiTTS Synthesis Streaming TTS Synthesis
NVIDIA · Shanghai, China
ReGen: Hierarchical Multi-Prompt Representation Generation for Efficient Waveform Diffusion Models
Jul 10, 2026Sang-Hoon Lee, Ha-Yeong ChoiAudio Representation Learning Latent Diffusion Models
Department of Artificial Intelligence, Ajou University, Suwon, Korea · KT Corp., Seoul, Korea
Towards Digital Preservation of Efik: TTS for a Low-Resource African Language
Jul 5, 2026Offiong Bassey Edet, Emmanuel Oyo-Ita, Archibong Okon Archibong +2TTS Synthesis Speech Generation Evaluation
University of Cross River State, Nigeria · ML Collective · University of Calabar, Nigeria
Towards a Phonology-Informed Evaluation of Multilingual TTS
Jul 2, 2026Sneha Ray Barman, Neeraj Kumar Sharma, Shakuntala MahantaTTS Synthesis Speech Generation Evaluation
Centre for Linguistic Science & Technology, IIT Guwahati · Mehta Family School for Data Science & Artificial Intelligence, IIT Guwahati · Department of Humanities & Social Sciences, IIT Guwahati
Enhancing Flow Matching with A Unified Guidance Framework for Efficient and Robust Speech Synthesis
Jul 1, 2026Zuda Yu, Qianhui Xu, Ting Chen +5Flow Matching Efficient Inference
Zuoyebang, China
LuxEmo: Expressive Text-to-Speech Corpus for Luxembourgish
Jun 30, 2026Nina Hosseini-Kivanani, Sandipana DowerahSpeech Processing Low-Resource TTS Synthesis
Radio T´el´evisioun L¨etzebuerg (RTL) & University of Luxembourg · Esch-sur-Alzette, Luxembourg · Tallinn University of Technology +1
Is Natural Always Appropriate? Investigating Naturalness and Appropriateness Across Different Domains for TTS Evaluation
Jun 30, 2026Dominika Woszczyk, Andreas Triantafyllopoulos, Jura Miniota +2TTS Synthesis Speech Generation Evaluation
Iconic, United Kingdom · Technische Universität München, Germany · KTH Royal Institute of Technology, Sweden +1
HPRO: Hierarchical Progressive Reward Optimization via Preference Extraction for Emotional Text-to-Speech
Jun 26, 2026Sihang Nie, Xiaofen Xing, Rui Xing +5Reward Modeling Speech Prosody
South China University of Technology, China · Huya Inc., China · Tongyi Fun Team, Alibaba Group, China +1
Closing the Quality Gap in Low-Resource Text-to-Speech: LoRA Fine-Tuning of VoxCPM2 for Khmer and Korean
Jun 25, 2026Phannet Pov, Hyun Woo Park, Voneat Pen +3Low-Resource TTS Synthesis Low-Resource Language Processing
Department of Big Data, Chungbuk National University, Cheongju, South Korea · Institute of Digital Research and Innovation, Cambodia Academy of Digital Technology, Phnom Penh, Cambodia · Department of Management Information Systems, Chungbuk National University, Cheongju, South Korea +1
CrossAccent-TTS: Cross-Lingual Accent-Intensity Controllable Text-to-Speech via Disentangled Speaker and Accent Representations
Jun 24, 2026Ram Annamdevula, Ankit Tatawat, Ashishkumar P. Gudmalwar +2Disentangled Representation Learning TTS Synthesis
Media Analysis, Sony Research India
Sarashina2.2-TTS: Tackling Kanji Polyphony in Japanese Speech Generation via Data Scaling and Targeted Data Synthesis
Jun 24, 2026Lianbo Liu, Shiao Zhu, Kai Washizaki +10TTS Synthesis Data Augmentation
SB Intuitions
CN-NewsTTS Bench: a target-level automatic benchmark for raw-input Chinese news TTS pronunciation
Jun 23, 2026Shijun LuoSpeech Generation Evaluation TTS Evaluation
NetEase Cloud Music, Hangzhou, China

Oct 19, 2026J/K move · Enter open · S save

SEAM: Spectral Embedding Alignment and Merging for Low-Resource TTS

Walaa Othman, Baher Mohammad, Stamatios Lefkimmiatis, Ammar Ali

ITMO University · SPC RAS · MWS AI

Text-to-Speech Low-Resource TTS Synthesis Multilingual TTS Synthesis

Open paper AskPDF arXiv

Expanding the vocabulary of conditional text-to-speech (TTS) models typically requires extensive fine-tuning with paired audio, creating a major bottleneck for low-resource and cross-lingual adaptation. To address this, we introduce SEAM, a zero-audio initialization method that constructs embeddings for unseen tokens from known subtokens. Unlike standard pooling strategies that ignore sequential structure, SEAM applies a spectral filter to preserve temporal phonetic order and uses closed-form ridge regression to align the new embeddings with the pretrained conditioning manifold. Across three languages and two TTS architectures, SEAM approaches or matches baseline performance using up to four times less training data, reducing sample complexity by 75% compared to random initialization and mean pooling. Our approach provides a lightweight, reproducible, and theoretically grounded solution for vocabulary expansion in low-resource speech synthesis.