---
license: mit
datasets:
- BharathK333/VoxMorph-Dataset
language:
- en
base_model:
- ResembleAI/chatterbox
pipeline_tag: text-to-speech
tags:
- ICASSP
- Audio-to-Audio
- Zero-shot-tts
- Voice-Morphing
- Security
papers:
- https://huggingface.co/papers/2601.20883
---
# VoxMorph: Scalable Zero-shot Voice Identity Morphing via Disentangled Embeddings

<!-- [**Project Page**](https://vcbsl.github.io/VoxMorph/) | [**Paper**](https://huggingface.co/papers/2601.20883) | [**GitHub**](https://github.com/Bharath-K3/VoxMorph) | [**Demo**](https://huggingface.co/spaces/Bharath-K3/VoxMorph) | [**Dataset**](https://huggingface.co/datasets/BharathK333/VoxMorph-Dataset) -->
<!-- [![Conference](https://img.shields.io/badge/ICASSP-2026-blue)](https://2026.ieeeicassp.org/) -->
[![ICASSP Paper](https://img.shields.io/badge/ICASSP-Paper-blue)](https://ieeexplore.ieee.org/document/11462383)
[![arXiv](https://img.shields.io/badge/arXiv-Paper-red)](https://arxiv.org/abs/2601.20883) 
[![Project Page](https://img.shields.io/badge/Project_Page-GitHub.io-blue)](https://vcbsl.github.io/VoxMorph/) 
[![Demo](https://img.shields.io/badge/Demo-HuggingFace-orange)](https://huggingface.co/spaces/BharathK333/VoxMorph) 
[![Dataset](https://img.shields.io/badge/Dataset-HuggingFace-yellow)](https://huggingface.co/datasets/BharathK333/VoxMorph-Dataset) 
[![Model](https://img.shields.io/badge/Model-HuggingFace-green )](https://huggingface.co/BharathK333/VoxMorph-Models)
[![Paper HF](https://img.shields.io/badge/HuggingFace-Paper-red?logo=huggingface)](https://huggingface.co/papers/2601.20883)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)

VoxMorph is a zero-shot framework that produces high-fidelity voice morphs from as little as five seconds of audio per subject without model retraining. The method disentangles vocal traits into prosody and timbre embeddings, enabling fine-grained interpolation of speaking style and identity. These embeddings are fused via Spherical Linear Interpolation (Slerp) and synthesized using an autoregressive language model coupled with a Conditional Flow Matching network.

This repository hosts the official model checkpoints for **VoxMorph: Scalable Zero-shot Voice Identity Morphing via Disentangled Embeddings** (ICASSP 2026). It contains the checkpoint files (`s3gen.pt` and `t3_cfg.pt`) for VoxMorph, a zero-shot TTS framework built on top of Resemble AI's frozen Chatterbox-TTS backbone.

## Citation

If you find this work useful in your research, please consider citing the ICASSP 2026 paper:

```bibtex
@INPROCEEDINGS{11462383,
  author={Krishnamurthy, Bharath and Rattani, Ajita},
  booktitle={ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, 
  title={VoxMorph: Scalable Zero-Shot Voice Identity Morphing via Disentangled Embeddings}, 
  year={2026},
  volume={},
  number={},
  pages={13332-13336},
  keywords={Filtering;Filters;Deepfakes;Vocoders;Videos;Protocols;HTTP;Wide area networks;Communication equipment;Communication systems;Voice morphing;text-to-speech;zero-shot learning;speaker embedding;interpolation;speech synthesis},
  doi={10.1109/ICASSP55912.2026.11462383}
}

@article{krishnamurthy2026voxmorph_arxiv,
  title={VoxMorph: Scalable Zero-Shot Voice Identity Morphing via Disentangled Embeddings},
  author={Krishnamurthy, Bharath and Rattani, Ajita},
  journal={arXiv preprint arXiv:2601.20883},
  year={2026}
}
```