Docsoft Releases Latest Version of Automatic Captioning System

Share Article

Docsoft, Inc., http://www.docsoft.com, is pleased to announce Docsoft: AV 2.0, the latest version of its audio captioning and transcription solution. The appliance, Docsoft:AV, has been developed to 'mine' or 'extract' audio content contained in digital audio and video files. Upon media capture, the content is converted to an XML-based file. This XML file is then indexed by means of a second interface, the Docsoft:SA search appliance.

Docsoft, Inc., http://www.docsoft.com, is pleased to announce Docsoft: AV 2.0, the latest version of its audio captioning and transcription solution. The appliance, Docsoft:AV, has been developed to 'mine' or 'extract' audio content contained in digital audio and video files. Upon media capture, the content is converted to an XML-based file. This XML file is then indexed by means of a second interface, the Docsoft:SA search appliance.

This XML-based transcript represents Docsoft:AV's 'listening' capability of the mined audio or video file. Accuracy of the speech-to-text conversion is a major limiting factor of similar speech recognition products. Docsoft has gone to great lengths to improve the accuracy of Docsoft:AV 2.0, thus reducing the user's valuable time required to correct missed words or pronunciations.

A successful media transcription will begin with a clear, broadcast quality audio file saved as a Windows Media, Apple Quicktime, MP3, MP4, Wav or other file format. Frequently, this high quality audio file will not be available and thus the transcript results obtained may result in a less accurate transcript. Several solutions to alleviate problems such as background noise, foreign or regional accents, and low quality audio capture can now be obtained from the new Docsoft:AV 2.0.

All interaction with the appliance is accessed and controlled via the intuitive Web interface. Speaker Profiles can be created that are specifically tailored to an individual speaker or subject matter. An exact transcript from the speaker, even relevant documents specific to the speaker's vocabulary or subject matter will generate higher precision in the conversion process. Once established, the Speaker Profile feature will train the system, thereby dramatically increasing the accuracy of the transcribed text for subsequent processed files.

Docsoft believes its products now can effectively close the gap in the digital audio file mining, searching and captioning markets. A key objective has been to reduce the cost and time formerly required to manage such media. "Our goal is to continuously implement improvements and features to the Docsoft:AV appliance," remarked Jacob Butcher, Docsoft President. Butcher continues, "There is a clear and obvious need in the educational, broadcasting, advertising and journalism fields for a product that can accurately manage digital media."

With increased emphasis now on 508 accessibility requirements, the Docsoft:AV and Docsoft:SA appliance solution give way to a new transformed process of transcription and closed captioning. The next version scheduled for release in late August contains a transcript editing component to streamline the audio to text conversion process.

About Docsoft:
Docsoft Inc. based in Oklahoma City is an XML industry leader. The company has been serving XML users by providing customized products and development services since 1999. Adding to existing XML expertise, Docsoft has evolved into the audio captioning marketplace with automatic speech recognition, a powerful tool that enables profound new search capabilities.

###

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Michael Robichaux
Visit website