IMS Toucan

IMS Toucan is an open-source, highly versatile text-to-speech (TTS) toolkit developed by the Institute for Natural Language Processing (IMS) at the University of Stuttgart. Built on a clean PyTorch framework, it is designed for teaching, research, and production speech synthesis, supporting advanced capabilities such as zero-shot voice cloning, expressive style transfer, and human-in-the-loop prosody editing (including pitch shifting and speed control). What uniquely distinguishes IMS Toucan is its expansive architectural design—leveraging articulatory phoneme representations and language-agnostic phonetic mapping—which enables high-quality speech generation and cross-lingual transfer across over 7,000 languages, making it a powerful utility for working with low-resource or endangered languages. Distributed under the MIT License, the platform provides developers and researchers with pre-trained model checkpoints, modular training pipelines, and interactive Gradio web interfaces to build self-hosted, scalable, and privacy-compliant voice applications.