Thursday, August 19, 2010

Nexiwave 2.5: Speaker Identiifcation

A quick announcement is the introduction is Nexiwave 2.5. We introduced a nice feature called Speaker Identification. Nexiwave automatically recognizes unique speakers in your audio files. Nexiwave can also recognize speakers across your audio files.

For more details, please check the "querySpeakEvents" and "listSpeakers" functions in our API documentation.

Monday, June 21, 2010

nexiwave featured in NVIDIA's CUDA Review

nexiwave's CUDA based Speech Indexing solution is featured in NVIDIA's latest CUDA Review. To recap, with a total rewrite of acoustic scoring, we moved more than 75% of computation into CUDA 3.0. Our further tuning achieved more than 75 times speedup.

Since the release of CUDA based solution into our production environment, we have been thinking really hard on both technical front, as well as on marketing directions. On the technical side, we are quite interested in moving the remaining computing cycles into CUDA hardware. However, as I explained in my other post, we realized that speed is just one side of the benefit. The very nature of massive computing power, as well as massive memory available, has made certain carefully crafted algorithm obsolete. More things are certainly cooking in nexiwave. More news to come on this front soon.

The whole CUDA Review can be found here:
http://www.nvidia.com/content/newsletters/web/CUDA_Week_in_Review_June_18_10.html

Friday, June 4, 2010

We are featured on GPGPU.org

The folks at GPGPU.org has been nice enough to feature us on their frontpage (so far, at least): http://gpgpu.org

A big thank-you from nexiwave!

Saturday, May 22, 2010

Annoucing GPU Powered nexiwave 2.0

We are proudly announcing that nexiwave has completed the GPU component of nexiwave's Speech Indexing service. nexiwave is gradually migrated its decoding clusters to a GPU centric cluster, due to the great cost-saving effect.

This is after over six months of planning, coding and testing. The speed improvement is nearly 75% relative to previous speed.

For more information, contact us at info@nexiwave.com

Tuesday, April 27, 2010

Automatic Media Uploader is now available

We have published a tool called "Automatic Media Uploader". The tool is quite straightforward, we think. It simply:
  • monitors a local dictionary every 10 minutes.
  • uploads any media files found in the directory (if they have not been uploaded yet.
It is still beta. It can be downloaded from:
http://www.nexiwave.com/static/downloads/AutomaticMediaUploader.zip

Audio Search API v2

We have released version 2.0 of our Audio Search API. Highlights are:
  • Post local files as SOAP attachment. This is much needed, since not everyone knows or wants to setup a server.
  • "multi" version of nearly all requests. All requests are now equipped with a batch version. For example, instead of adding files one by one, there is multiAddExt, which can be used to add (or upload) many audio/media files at once.
There were quite some other improvements. For more details, access the api doc here: http://www.nexiwave.com/PC-NG-AA/api/nexiwave.audio.search.SaaS.api.pdf

Saturday, December 12, 2009

Audio Search SaaS API published

We have published our Audio Search SaaS API. In nutshell, these are the API calls:
  • add: send audio/media to us
  • search
  • getTranscription
  • delete
  • lookup
For general info, visit: http://nexiwave.com
For detailed API document: http://www.nexiwave.com/PC-NG-AA/api/nexiwave.audio.search.SaaS.api.pdf
For wsdl file: http://search.nexiwave.com//AudioSearch/services/SearchService?wsdl