Powered by Universal Speech Solutions LLC


Kaldi Speech Recognition Plugin 1.0.0 Released

Kaldi Speech Recognition (KaldiSR) Plugin 1.0.0 to the UniMRCP Server (UMS) has been released.

The plugin is based on the following components:

  • UniMRCP Server 1.5.0
  • Kaldi GStreamer Server
  • Libevent 2.1.8
  • Rapidjson 1.1.0

The binaries are currently available for the following Linux distributions:

  • Red Hat / CentOS 7
  • Ubuntu 16.04 LTS

Kaldi Speech Recognition Toolkit can now be used by IVR platforms via MRCP.

The Kaldi plugin to the UniMRCP server connects to the Kaldi GStreamer Server, which needs to be installed separately. This integration is primarily intended for dev teams experienced with Kaldi building their own speech recognition systems with a special attention to Deep Neural Networks (DNNs). The plugin allows both an easy integration and reuse of existing infrastructure.

For more information on the toolkit and the GStreamer server, please check out the following resources:

Special thanks are extended specifically to Yinfeng Qin for his help and knowledge in the Kaldi infrastructure, and also to the maintainers of the respective projects listed above.

For installation, configuration and usage instructions, consult the guides available on a new Kaldi SR plugin page.


Please note this is a commercial module. Everyone is welcome to start evaluation by requesting a trial license. The production license is priced at $50 per channel.

Thank you for using UniMRCP.

Arsen Chaloyan
Author of UniMRCP

Announcement Group

Subscribe to the Announcements Group if you prefer to receive news and press releases only. Please note that announcements are normally posted to the Discussion Group as well.