nexiwave.com, a Speech Indexing Cloud Service company based in Boston MA, announces that it has completed the GPU-acceleration of its speech indexing service, Nexiwave 2.0. Without sacrificing accuracy of its service, nexiwave enjoys over 75% relative speed improvement (comparing a stock Sphinx4 running on a 2.5Ghz/8 core/24GB RAM server to a Sphinx4 on 2.5Ghz/Quad Core/4GB with NVIDIA GTX 470 GPU).

Nexiwave’s Speech Indexing system is built on top of the CMU Sphinx4 open-source speech engine, with certain proprietary improvements. Nexiwave’s focus market is effective information retrieval from audio archives. The Nexiwave system is tailored towards large vocabulary, speaker independent conversational audio. Nexiwave’s service is tailored for these industries:

• Keyword search for two way conversational telephony recordings
• Keyword search for multi-party conference call recordings
• Machine transcription for high quality consumer grade audio recordings

Nexiwave offers its product in both web service and product license form. http://nexiwave.com
Nexiwave  blogs: [http://blog.nexiwave.com, http://nsh.nexiwave.com, http://ben.nexiwave.com]