skip navigation

Docsoft, Incorporated

Audio Mining

The Docsoft:AV appliance is a powerful software and hardware solution designed to audio mine (capture) the spoken content in digital audio and video (AV) files. This appliance can be used to automatically generate text transcripts and closed captioning formats and can be configured to output virtually any text based format.

Docsoft:AV will enhance any captioning or transcription workflow, AV data mining, archiving, or search solutions. Docsoft:AV is a core component to the Docsoft AV Search solution and the Docsoft AV Captioning solution.

How it works

Docsoft:AV uses an advanced speech recognition engine to analyze the spoken content within digital AV files and converts it to an XML based format. The result can automatically be transformed to output formats such as Quicktime Text, RealText, SAMI or plain text.

Combining the generated transcript file with the video will allow users to create the necessary closed caption feature for ADA compliance. Using the Docsoft:SA, indexing and search capabilities can now be obtained from the transcript originally generated by the Docsoft:AV appliance.

Improving Speech Recognition

At the user’s option, speaker profiles can be created to improve the speech recognition of the system for a specific speaker. The speaker profile can be trained to the speaker's voice and words can be added to the speaker's vocabulary. Training will increase speech recognition accuracy of output.

Training the speaker profile is a simple process that does not require the speaker to interact with the system.

Docsoft:AV Accessibility

To learn how Docsoft:AV fits in with your accessibility requirements, please view the following :

Online Demonstration

If you would like an online demonstration of Docsoft's automatic speech to text capabilities, please contact us and ask for a demonstration of this powerful technology.

Online Video Training

Access the training library by clicking here.