Speaker Disentanglement

Follow this topic Ask about this topic

Momentum

1 paper in the last four weeks, with none the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 10

All topics

DEFINE: Exemplar-Guided Accent Control for Zero-Shot TTS
Sep 26, 2026Ambuj Mehrish, Abhinaba Roy, Alex Ivanov +2TTS Synthesis Zero-Shot TTS
Ca’ Foscari University of Venice · Singapore University of Technology and Design · Kandinsky Lab +1
Content is What Remains: Invariant Speech Tokenization from Parallel Utterances
Jul 21, 2026Laurin Wagner, Bernhard Thallinger, Miroslav Stankovic +1Audio Tokenization Self-Supervised Speech Representation Learning
nyra labs · nyra labs, Austria
Speaker-Disentangled Chunk-Wise Regression for Syllabic Tokenization
Jul 5, 2026Ryota Komatsu, Kota Kawakita, Takuma Okamoto +1Audio Tokenization Self-Supervised Speech Representation Learning
Institute of Science Tokyo, Meguro, Tokyo 152-8550, Japan · National Institute of Information and Communications Technology, Kyoto 619-0289, Japan
Beyond Cross-Reconstruction: Probing-Based Disentanglement Evaluation for Acoustic Teleportation Codecs
Jun 30, 2026Philipp Grundhuber, Emanuël A. P. HabetsDisentangled Representation Learning Neural Audio Codecs
Fraunhofer Institute for Integrated Circuits (IIS), Erlangen, Germany · International Audio Laboratories Erlangen∗, Erlangen, Germany
SDP-Codec: A Speaker-Decoupled Speech Codec with Pitch Injection for Low-Bitrate Coding and Zero-Shot Voice Conversion
Jun 19, 2026Hounsu Kim, Juhan NamNeural Audio Codecs Speech Processing
Graduate School of Culture Technology, KAIST, Daejeon, South Korea
Dual-Granularity Orthogonal Disentanglement for Generalizable Audio Deepfake Detection
Jun 15, 2026Zhuodong Liu, Hugen Lv, Xiangyu Li +1Representation Learning Audio Deepfake Detection
Beijing Jiaotong University, China · Shanghai Jiao Tong University, China · ITMO University, Russia
Multilingual Multi-Speaker Unit Vocoders: A Systematic Analysis of Discrete Speech Representations
Jun 4, 2026Naman Kothari, Arjun Gangwar, Adarsh Arigala +1TTS Synthesis Speech Processing
National Institute of Technology, Trichy · Indian Institute of Technology, Madras
Speaker-Disentangled Remote Speech Detection of Asthma and COPD Exacerbations
May 16, 2026Yuyang Yan, Sami O. Simons, Visara UroviSpeech Processing Clinical Prediction
Institute of Data Science, Maastricht University, Paul-Henri Spaaklaan 1, Maastricht, 6229 EN, the Netherlands · Department of Respiratory Medicine, NUTRIM Research Institute of Nutrition and Translational Research in Metabolism, Faculty of Health Medicine and Life Sciences, Maastricht University, P. Debyelaan 25, Maastricht, 6229 HX, the Netherlands · Department of Respiratory Medicine, Maastricht University Medical Centre, P. Debyelaan 25, Maastricht, 6229 HX, the Netherlands
Mixed-Precision Information Bottlenecks for On-Device Trait-State Disentanglement in Bipolar Agitation Detection
May 4, 2026Joydeep ChandraMixed-Precision Quantization Information Bottleneck
Department of Computer Science & Technology, Tsinghua University, Beijing, China.
Unified Architecture and Unsupervised Speech Disentanglement for Speaker Embedding-Free Enrollment in Personalized Speech Enhancement
May 18, 2025Ziling Huang, Haixin Guan, Yanhua LongSpeech Enhancement Speaker Disentanglement
Shanghai Engineering Research Center of Intelligent Education and Bigdata, Shanghai Normal University, Shanghai, 200234, China · SHNU-Unisound Natural Human-Computer Interaction Lab, Shanghai Normal University · Unisound AI Technology Co., Ltd., Beijing, China

Sep 26, 2026, cs.SDJ/K move · Enter open · S save

DEFINE: Exemplar-Guided Accent Control for Zero-Shot TTS

Ambuj Mehrish, Abhinaba Roy, Alex Ivanov, Tawsif Ahmed+1

Ca’ Foscari University of Venice · Singapore University of Technology and Design · Kandinsky Lab+1

TTS Synthesis Zero-Shot TTS Speaker Disentanglement

Open paper AskPDF arXiv

Zero-shot text-to-speech (TTS) can reproduce an unseen speaker from a short reference recording, but typically entangles speaker identity and accent within the same reference. We introduce DEFINE, an end-to-end framework that decouples these factors by conditioning speaker identity and target accent on separate audio exemplars. A single inference-time guidance weight continuously controls accent strength without retraining. Built on F5-TTS with parameter-efficient LoRA adaptation, DEFINE maps short accent exemplars into a conditioning space using an exemplar encoder supervised through learned accent prototypes, requiring neither accent labels at inference time nor post-synthesis waveform conversion. On seen accents, increasing accent guidance improves accent-probe accuracy from 6.5% to 19.6%. More importantly, a single DEFINE model generalizes accent control beyond its training accent set: on seen and out-of-domain accents, though not on held-out accents, it matches the accent transfer performance of a two-model TTS-voice-conversion cascade while achieving higher speaker similarity and comparable predicted speech quality. These results demonstrate that speaker identity and accent can be independently controlled from audio exemplars within a single zero-shot TTS model, including for accents unseen during training.

Momentum

Related topics

Latest papers 10