What is Time-Scale-Modification?

Time-Scale Modification (TSM) refers to the process of compressing or expanding the Time-Scale of an audio segment. A signal which is time-scale compressed has a shorter duration, while a Time-Scale expanded signal is longer in duration. Simply speeding up the playback rate of a digital signal resampling causes the local pitch periods to be shortened. This shortening increases the frequency and the resulting signal sound like chipmunks. A signal created this way is has a shorter Time-Scale, but is largely unintelligible.

A properly Time-Scale Modified signal maintains properties of the original signal such as local pitch period, speaker identity, and intelligibility. It does so by preserving the prominent features of the signal associated with these properties, namely the local pitch period. When this is performed on a voice signal, the resulting signal sounds as though the same person is talking faster or slower in the same voice.

Time-Scale Modification is a powerful technique because it allows listeners to throttle the rate of audio-information in much the same way a reader controls his/her reading rate by moving his/her eyes across a page.

Applications of Time-Scale Modification

There are a large number of applications in which it is desirable to modify the time-scale of speech, music or other acoustic material without modifying the pitch. Radio stations can use the technique to speed up dance music, audio passages may be fast-scanned, the blind can "speed-read" audio lectures, instructional material for foreign languages may be slowed down to aid in comprehension. But that's not all.....

Digitized speech as an information medium is ubiquitous on current computers, the World-Wide-Web, Compact Disc Players, and "Talking Books". The digital audio medium will soon enter our automobiles and homes for news broadcasting, reporting, and entertainment.

Consider the applications:

• Voice-Mail Messaging
• Adjustable Voice Prompt Speeds
• Talking-Books
• Web-based Audio Browsing
• Broadcast Radio
• Video replay at any speed with full audio accompaniment
• Voice-Memo Systems
• Court Reporting Training
• Transcription Training
• Transcription Services
• Voice Logging
• Foreign Language Learning
• Speed Reading for the Blind
• Compressing Commercials into convenient time-slots without deletion
• Works in ANY LANGUAGE.