Time-assigned language interpolation

from Wikipedia, the free encyclopedia

At the time assigned speech interpolation ( ZSI , English time-assignment speech interpolation , TASI ) is one of Bell Labs developed dynamic multiplexing method for analog audio signals, in which the participants only one of the channels of a transmission medium is allocated when actually spoken.

The participants are only assigned a channel as long as audio signals are registered. If there is a pause in speech, the channel is released immediately so that it can be used by another subscriber connection. If speaking is resumed, another, currently free channel is allocated. The new channel is also released again when there is silence. In this way, each user channel of the transmission medium can be used quasi simultaneously by several subscriber connections.

This method requires a central management channel for signaling . In this management channel, the transmitter constantly informs the receiver of the current channel occupancy of the transmission medium. With this information, the receiver can then reassemble the individual time segments of a subscriber connection, which were split up and transmitted over several channels, to form a continuous signal.

With this method, the switchings result in a loss of speech information. If the switchover can be done very quickly, the voice quality is not significantly reduced by these losses. Because of the lack of synchronization between sender and receiver, this dynamic multiplex process is called "asynchronous".

First practical application

The second transatlantic cable, TAT-2, went into operation on September 22, 1959. In the spring of 1960, the number of speech channels was increased from an initial 47 to 87 using the TASI process.

literature

  • Franz Miller: The mp3 story. A German success story, Carl Hanser Verlag, Munich 2015, ISBN 978-3-446-44471-3 .
  • Patent US5016247 : Multihop time assigned speech interpolation (TASI). ‌