The Cambridge University Python Multi-domain Statistical Dialogue System Toolkit

PyDial is an open-source end-to-end statistical spoken dialogue system toolkit which provides implementations of statistical approaches for all dialogue system modules. Moreover, it has been extended to provide multi-domain conversational functionality. It offers easy configuration, easy extensibility, and domain-independent implementations of the respective dialogue system modules.

PyDial is a scientific system developed for scientific purposes and actively used for research by the Dialogue Systems Group at the University of Cambridge. Although we put great effort into designing clean interfaces and easily understandable structures, some the code will not necessarily be up to commercial standards. As it is used actively by our research group, the code may also change in the future. However, the core interfaces have been designed in a way that there should not be any need to change them.

For more information and code downloads, visit


PyDial RL-based DM Benchmarking environment

Reproducibility has been an open issue in RL-based dialogue management for several years. In order to mitigate this issue, we have implemented in PyDial a set of benchmarking simulated dialogue environments which span different domains, user behaviours and input noise conditions. For more information check the paper and the PyDial benchmarks webpage.