Speech-to-text perfected

Digital Syphon Sonic Imagën

OCR to text, handwriting to text, Logo identification.

With Sonic Imagën you can perform simultaneous OCR processes against a corpus of image based documents. You can perform logo searching or identification, and facial detection from images or thumbnails from video sources.

All one has to do is direct Sonic Imagën to the appropriate Folder path to your target files for OCR, logo or facial recognition. For logos and faces, one has to simply provide a source location of the training material or items to match by filename.

Sonic Imagen, is a highly flexibly, powerful OCR software that enables image to text and logo detection to happen in a multi process fashion. Imagine having many different scanned images in various file types like PNG, JPG, GIF, TIFF, PDF and knowing that Sonic Imagën is capable of extracting the text from these scanned documents and generating a text and xml file of the documents, all the while performing this 2-8x faster than any other product out there. The engine can take in various file types and the content can be portrait and landscape views, legal, letter or other file sizes and the system will automatically detect each and every type thereby saving you the time and energy of having to process homogenous document layouts of the same type as in other OCR systems. In addition the engine also has the capability to match logos from a repository of files that need to be matched according to what logo to recognize and generate a report based on this. The engine is capable of being configured to OCR most European, and Asian languages depending on the need.

Sonic Imagen is a lightweight engine that operates on windows/linux OS and allows easy integration with a Java SDK. Most OCR systems are not able to process VECTORED images, or images in which are all based on mathematical expressions, and not Bitmap images like those found commonly in BMP type image formats. Sonic Imagen handles both, and like many Engines by Digital Syphon one can run in configurations of 2,4 or up to 8 processes at a time, meaning you can process multiple files at the same time and generate text based output, thereby allowing one to process thousands of scanned image documents in minutes rather than hours or days. In addition to the qualities mentioned, Sonic Imagen has the capability to identify logos that need to be identified against a large repository of images. This allows one to be able to detect a political action groups, SUPERPAC logos or simply a logo like the Pepsi brand in a particular font style against many thousands of images in which one doesn’t’ know if it’s located in there or not. The data that is generated is then a list of matches in a text file that highlights the path location and name of the file matched against what logo if there is a match. This logo/face detection is dynamic in that the image need not be exactly matching in size, for example one could have a small logo the size of 50x50 pixels and if the images being searched against contain a logo that is 70x70 pixels the system would detect it as a likely match.



Benefits:
  • Obtaion a powerful ROI immediately with Sonic Imagën
  • View, Process, Read for productivity
  • No end user agents to install
  • Easily adaptable to any environment
  • Internationalize for all major languages



To speak with a Digital Syphon Sales Manager, please contact us at 469-854-3528.
Contact our Sales team by email at sales@digitalsyphon.com