As of today, CMUSphinx project is proud to present you the Spanish acoustic model. It was trained on the acoustic data to support both wideband recognition of microphone speech recordings and narrowband recognition of the telephone speech. Now you can build Spanish IVR using CMUSphinx tools.
You can find the model HERE.
Please try and report the accuracy results.
Also, note that there are models for many languages like German, Dutch, French or Russian. In the future we are going to expand the set of languages for your convenience.