Latest Text-to-Speech Research Papers

The newest Text-to-Speech papers from across the field — arXiv, NeurIPS, CVPR, Nature, and more — refreshed daily and ranked by relevance. Distill AI tracks Text-to-Speech so you don’t have to: get the standout work delivered to your inbox every morning, with 2-sentence summaries and the option to chat with any paper.

Recent papers

• Discovering Interpretable Acoustic Features in Autoregressive Voice-Cloning TTS Models
  Anuj Diwan, David Harwath, Eunsol Choi · arXiv · Oct 29, 2026

  Voice-cloning TTS models synthesize an unseen speaker's voice from brief speech prompts, but it remains unclear which acoustic properties they represent and use. We introduce a feature-based interpretability framework for autoregressive voi…

• What Makes a Good Audio Prompt for Zero-Shot TTS? Learning Prompt Selection from Synthesis Outcomes
  Arnon Turetzky, Yossi Adi · arXiv · Oct 29, 2026

  Zero-shot text-to-speech (TTS) models clone a target voice from a short reference clip, commonly referred to as the prompt. In practice, the prompt has to be cut from a longer recording of the speaker, so which part to use must be chosen. T…

• Toward Complete Speaker Anonymization: a Streaming Cascaded System for Voice Privacy and PII Removal
  Stanisław Kacprzak, Thomas Thebaud, Najim Dehak · arXiv · Oct 29, 2026

  We present a cascade streaming voice privacy system, consisting of ASR, NLP, and TTS modules. By leveraging recent improvements in all three aspects of speech and language processing, we show that such a system is capable of achieving a res…

• EmpDPO: Enabling Emphasis Control in TTS via Direct Preference Optimization
  Haonan Chen, Yunyun Wang, Megan Wei, Zeyu Jin · arXiv · Oct 29, 2026

  Word-level emphasis conveys meaning that text alone cannot express and is therefore essential for a practical text-to-speech (TTS) system. In this work, we study how to add emphasis control post hoc to a pretrained TTS model that has never …

• Reading-Level Verification of Kanji Words in Japanese Financial Speech Synthesis
  Cheng Zhang · arXiv · Oct 29, 2026

  Few-shot voice cloning makes it cheap to produce daily Japanese financial commentary with several synthetic speakers, but kanji words with more than one reading, such as the word for "market" (shijo vs. ichiba) or "overall trend" (taisei vs…

• ConformalMOS: Uncertainty-Aware MOS Prediction with Conformal Intervals and Ordinal Modeling
  Kehinde Elelu, Joshua E. Siegel, Mohammadali Saffary, Tashfain Ahmed et al. · arXiv · Oct 26, 2026

  Accurate prediction of human Mean Opinion Scores (MOS) is crucial for evaluating synthetic speech quality in text-to-speech (TTS) and voice conversion (VC). Most MOS predictors output only point estimates and ignore uncertainty, limiting re…

• When the Persona Contradicts the Line: Diagnosing Plan-to-Synthesis Failures in Chain-of-Thought TTS
  Joonyong Park, Jerry Li · arXiv · Oct 26, 2026

  Chain-of-thought TTS systems rely on language-model planners to translate persona and context into vocal delivery, but it is unclear where this intent is lost when the requested affect conflicts with lexical sentiment. We introduce a biling…

• ImageTTS: Repurposing Pretrained Image Diffusion Transformers for Speech Synthesis
  Hanchen Pei, Shujie Liu, Yayue Deng, Fangyun Wei et al. · arXiv · Oct 26, 2026

  Pretrained image-generation models have been successfully repurposed for various downstream visual tasks. Whether their learned representations and generative priors can transfer beyond vision to other modalities remains largely unresolved,…

• LeanTTS: Efficient Full-Stack Bilingual TTS for Zero-Shot Cloning and Voice Customization
  Yuheng Lu, Qiulin Li, Xi Wang, Yang Wang et al. · arXiv · Oct 25, 2026

  While recent advances in text-to-speech (TTS) synthesis have substantially improved speech quality and modeling flexibility, the development of lightweight and computationally efficient systems remains comparatively underexplored. We presen…

• DRAMA-Eval: A Benchmark for Context-Aware, Multi-Turn, Multi-Speaker Speech Generation
  Jingbin Hu, Zihang Yang, Xiaoge Peng, Haoyu Zhang et al. · arXiv · Oct 22, 2026

  Recent advances in speech generation are shifting toward context-aware, multi-turn, and multi-speaker synthesis. While these capabilities are in high demand for applications such as audiobook and audio drama production, evaluation has not k…

• Measuring the Verifier's Floor: How Much Apparent Rare-Word Text-to-Speech Error Is ASR Error?
  Kirill Borodin, Vasilii Kudryavtsev, Nikita Vasiliev, Grach Mkrtchian · arXiv · Oct 22, 2026

  Text-to-speech (TTS) is usually evaluated by transcribing its output with an automatic speech recognition (ASR) system, so the word error rate also counts recognizer errors. We measure these errors on human recordings of the test sentences.…

• EmoFI-TTS: Face-Conditioned Expressive Voice Synthesis with Intensity Modulation
  Yejin Jeon, Luel Hagos Beyene, Marie Maltais, Hyounghun Kim et al. · arXiv · Oct 21, 2026

  Controlling emotional intensity remains challenging in expressive text-to-speech synthesis, particularly when emotion must be inferred without explicit labels. We propose EmoFI-TTS, a face-conditioned framework that derives emotion category…

• HeyGen Voice: Scene-Aware Voice Cloning for Talking Videos
  Pavel Andreev, Andrei Gusev, Adithya Narayan, Yi Ren et al. · arXiv · Oct 21, 2026

  Most voice-cloning systems are built for audio alone: they copy the reference recording's delivery and acoustics, so a voice cloned on a phone sounds like a phone recording whether the speaker is pictured in a studio, on a street, or in a c…

• Decodability Is Not Controllability: Local and Anisotropic Speaker Control in Continuous-Latent TTS
  Yuanzhong Zheng, Wei Huang, Ming Ma, Kailun Jin et al. · arXiv · Oct 21, 2026

  High linear decodability does not entail causal controllability in continuous-latent text-to-speech (TTS). In VoxCPM2, position-matched probes and equal-norm interventions across 30 speakers and 60 directed dyads expose this separation: ref…

• Where Does Timbre Hide? Localizing Speaker Information in Speech Signals
  Yiquan Zhou, Yixuan Wang, Xiaozhe Qi, Xin Gao et al. · arXiv · Oct 21, 2026

  Timbre similarity is a key evaluation target for generative speech systems, including voice conversion, speaker-cloning text-to-speech, and zero-shot speech models. When the generated voice does not match the target speaker, the problem oft…

• Character-Level Prosody Control for Mandarin Speech Synthesis:An Architecture-Preserving Approach
  Diankun An, Lu Ma, Yongliang Wang, Zhewen Su et al. · arXiv · Oct 21, 2026

  This paper addresses fine-grained prosody control for large-language-model (LLM) based text-to-speech (TTS) synthesis of Mandarin Chinese, motivated by multimedia content-creation scenarios---film dubbing, audiobook production, and interact…

• LoCA: Local Credit Assignment for Reinforcement Learning of Autoregressive Flow-Matching TTS
  Jie Liu, Junyu Zhou, Junhao Zhong, WeiSen Tang et al. · arXiv · Oct 20, 2026

  Reinforcement learning can improve text-to-speech synthesis through feedback on generated speech. However, sentence-level rewards do not identify local errors, while signed regression updates can lead to excessive fitting or repulsion of sa…

• Steering intonational tunes in text-to-speech synthesis
  Juan Pablo Vigneaux, Jennifer Cole · arXiv · Oct 19, 2026

  Text-to-speech systems built on residual vector quantization often generate audio in two stages: a transformer backbone emits the first level of each frame's token stack, and a smaller depth model generates the rest autoregressively. We ask…

• MixiSpeech: Progressive State Distillation for One-Step Vietnamese TTS
  Tran Quoc Bao Le, Binh Mai, Hung Dinh, Lam Nguyen et al. · arXiv · Oct 18, 2026

  Conditional-flow text-to-speech (TTS) models can achieve high synthesis quality with relatively few sampling steps, but collapsing the remaining trajectory to a single update can still cause substantial quality degradation. We propose MixiS…

• Does Expert Supervision Improve Text-to-Speech? An Empirical Study of Preference Optimization
  Minh Duc Vu, Junchuan Zhao, Ye Wang · arXiv · Oct 18, 2026

  Preference optimization promises to improve text-to-speech (TTS) systems by learning which speech outputs are better. External expert speech seems a natural source of preferred responses, especially when student generations contain errors. …

• ShlokaDhara: A Multi-Speaker Sanskrit Speech Corpus for Learning Meter-Aware Prosody in Text-to-Recitation
  Harshal Kausadikar, Onkar Kishor Susladkar, Prathmesh Santosh Choudhari, Vedangi Deepak Deshpande et al. · arXiv · Oct 17, 2026

  Sanskrit verse recitation is governed by $\textit{Chandas}$, a strict prosodic framework defined by rhythmic $\textit{Laghu}$ (light) and $\textit{Guru}$ (heavy) syllables. Because existing Sanskrit speech datasets prioritize ASR or convers…

• SENTRY: A Streaming Edge Network Trust Engine for Real-Time Audio Deepfake Detection
  Chirantan Banerjee, Soumita Chatterjee, Bhargab B. Bhattacharya, Hafizur Rahaman · arXiv · Oct 17, 2026

  Real-time voice cloning has enabled large-scale streaming deepfake and vishing attacks, while existing Network Detection and Response (NDR) solutions rely on GPU-based post-processing that requires buffered audio and generalizes poorly to u…

• Beyond the Benchmark: Conditioning, Domains and Failure Modes of Qwen3-TTS
  Séverin Baroudi, Yanis Labrak, Petr Motlicek, Hervé Bredin et al. · arXiv · Oct 16, 2026

  Qwen3-TTS reports state-of-the-art zero-shot voice cloning, but its technical report leaves open how the model should be conditioned, how it behaves outside read speech, and how it fails. We characterise its two cloning interfaces, in-conte…

• DriftTTS: One-stage Single Step Text-to-Speech via Generative Drifting
  Jaehyun Ahn, Dohyun Kim, Heeseung Kim · arXiv · Oct 16, 2026

  Text-to-speech models built on score matching and flow matching produce natural speech, but only by calling the decoder many times, which makes synthesis slow. Existing ways to cut the number of calls either distill from a multi-step teache…

• Too Emotional to Be Fake? Intra-Utterance Emotion Shifts Expose Failure Modes in Speech Deepfake Detectors
  Jingyi Chen, Surya Koppisetti · arXiv · Oct 16, 2026

  Expressive speech synthesis can now change emotion within a single utterance, yet its impact on deepfake detection remains unclear. We introduce EmoShift, a benchmark comparing emotion-stable (AA) and emotion-changing (AB) speech to examine…

• Noise-Aware Lombard Speech Synthesis via Flow-GRPO for Intelligibility Optimization
  Zhaoyang Wang, Dingding Yao, Chengzhong Wang, Junfeng Li · arXiv · Oct 14, 2026

  Voice-based conversational agents are increasingly deployed in real-world settings, where environmental noise often degrades the interactive experience. Therefore, robust acoustic adaptability is essential to maintain natural and intelligib…

• Long-Prompt Zero-Shot TTS: Benchmarking Prosody Reproduction Beyond Voice Cloning
  Dekun Chen, Renjie Xu, Yingda Shen, Qinke Ni et al. · arXiv · Oct 14, 2026

  Zero-shot text-to-speech (TTS) typically conditions on a short speech prompt, which conveys timbre but provides only limited evidence of a speaker's prosodic patterns. We formulate Long-Prompt Zero-Shot TTS, a task that uses long speech pro…

• Speaking Longer Than Learned: Training-Free Long-Form Generation for Full-Stream Text-to-Speech
  Jeseung Lim, Jiseong Woo, Heeseung Kim · arXiv · Oct 14, 2026

  Recent full-stream text-to-speech (TTS) models enable low-latency personalized synthesis as input text arrives. However, sustaining generation beyond the lengths seen during training without additional training remains largely unexplored. W…

• Revise-TTS: A Non-Masking Refinement Framework for Masked Diffusion Models in Text-to-Speech
  Wanli Sun, Dan Su, Anton Ragni · arXiv · Oct 14, 2026

  Text-to-speech (TTS) based on masked diffusion models (MDMs) achieves competitive zero-shot speech synthesis performance. Unlike autoregressive (AR) models, MDMs define a corruption process that maps clean speech-token sequences to masked m…

• Beyond Fixed Targets: On-Policy One-Shot Speaker Personalization for Autoregressive Text-To-Speech Models
  Sangkwon Park, Che Hyun Lee, Heeseung Kim, Sungroh Yoon · arXiv · Oct 13, 2026

  One-shot personalization of autoregressive (AR) text-to-speech (TTS) models remains challenging because supervised fine-tuning (SFT) on a single reference pair rapidly degrades generation quality. We identify fixed, off-policy supervision a…