Audio and Speech Analysis

Autonomy solutions enable live or recorded speech to be manipulated, edited, searched and automatically processed in the same way as text. This is achieved through SoftSound, renamed Virage SoftSound in 2007, which provides a wide range of speech-processing technologies, from audio segmentation and identification through to automatic speech recognition and understanding. SoftSound was founded in 1995 and is backed by over 10 years of research at Cambridge University.

Autonomy Virage Courtroom Transcription

Key features include:

State-of-the-art accuracy

Support of phonetics, word spotting and conceptual indexing and search

Highly scalable real-time recognition

Virtually unlimited vocabulary size

Industry-leading range of supported languages

Audio processing capabilities:

Speaker / audio segmentation

Speaker / audio classification

Identification of:

Topics being discussed

Genders of the speakers

Emotional character of speech

Amount and location of speech versus non-speech (e.g. background noise or silence)

Linguistic origin of speakers

Music

Audio Conceptual Tracking

A true industry first, Autonomy's Audio Conceptual Tracking tools allow users to analyze all the information held in rich media assets in a single view. Using significant breakthroughs in expectation maximization algorithms and usability modelling, Autonomy enables users to have access to all the concepts that are represented at different points in the audio stream, even as they are focusing on a particular point of a clip. This is a particularly empowering function since it allows users to work with rich media assets on a holistic basis rather than myopically on the segment they may be listening to.

Another important innovation and a true paradigm shifter, has been the ability for users to dynamically control the confidence of the speech recognition engine to match the business requirements per query. In traditional systems that deal with speech, the user typically must query what is essentially a black box for an answer and has no control over the confidence setting of the speech processing technology in use. For the first time, Autonomy allows end users to change the configuration of the speech recognition engine on the fly, per query, to suit their business needs. The higher the confidence setting, the lower the number of returned results but with fewer number of false positives.

In sum, the key features include:

Intelligent, concept-based marking of key ideas

Expansion of the concept viewer to encompass the entire asset in a single view, or focus on narrow parts of the asset

User-controlled confidence configuration

Sound redaction interface

Further Reference:

Autonomy Virage Rich Media Management

Further Reference:

Autonomy etalk Contact Center Solutions

Functionality

PODS & Portlets

Audio & Speech Analysis

Video Analysis