Digital memory providing fixed and variable delays in a TASI system

• US 4,284,850 A
• Filed: 01/12/1979
• Issued: 08/18/1981
• Est. Priority Date: 12/23/1977
• Status: Expired due to Term

• Current Assignee: Cherry Treet Ventures I, Edelson Technology, Parsnip River Company, Republic Telcom System Corporation, Reynolds Creek Limited Partnership
• Original Assignee: Storage Technologies Corporation (Oracle Corporation)
• Inventors: Clingenpeel, Glenn R.
• Primary Examiner(s): Robinson, Thomas A.
• Application Number:
  US06/003,032
• Time in Patent Office: 949 Days
• Field of Search: 179/15 AS, 179/15 A, 179/15 BW, 179/15 BA, 179/15 BV, 179/15 AP, 179/15 BY, 179/1 SA
• US Class Current: 370/435
• CPC Class Codes: H04J 3/17 in which the transmission channel allotted to a first user may be taken away and re-allotted to a second user if the first user becomes inactive, e.g. TASI speech analysis or identification G10L
• CPC Class Codes: H04J 3/177 Freeze-out systems, e.g. taking away active sources from transmission

Abstract

A time assignment speech interpolation system has a random access memory which provides both a fixed and a variable delay between an input speech channel and a transmission facility. Digital samples of the signal detected on an input speech channel are stored for a fixed time in a portion of the memory in order to provide time for switching without clipping and for generation of a channel identifying symbol. If a transmission facility is available, the sample is read out of memory after the fixed delay and reconstituted into an analog signal which is transmitted on the available facility. If no transmission facility is available, the samples are read out of memory after the fixed delay and written into another portion of memory which stores them for variable times until a transmission facility is available. In one embodiment, the portions of the memory assigned to providing variable delay are paired. Starting addresses are at the two extremes of memory locations in a pair, and successive memory locations extend toward the other extreme starting address. In another embodiment, the memory assigned to providing variable delay is segmented. As one segment is filled, it is linked to another available segment. In this embodiment, silence purging logic may also be provided which determines whether any of the samples stored in particular segments represent silence or pauses in the stored speech. If they do, these segments are then classified as available.

Claims

1. A time assignment speech interpolation system comprising:

means for sampling signals detected on each input speech channel;

a random access digital memory connected between said input speech channels and transmission facilities extending to a remote location, said samples being stored in a fixed portion of said memory for a fixed time;

an address generator for generating addresses in a variable portion of said memory which stores said samples in linked segments of said variable portion;

control means having a first portion for storing the addresses of said samples in said linked segments of said variable portion for variable times until a transmission facility is available and subsequently for retrieving said samples in said linked segments for transmission to a transmission facility.

2. The system recited in claim 1 wherein said control means further comprises a second portion for storing a table of available addresses in said variable portion, said address generator being responsive to said second portion.

3. The system of claim 1 wherein said signals are detected on said input channels by means of speech detectors, said control means being responsive thereto and further comprising:

a logic circuit responsive to said control means for removing the addresses of those segments of said linked segments in which no samples have been stored from said first portion.

4. The system of claim 3 wherein the addresses of said removed segments are stored in said first portion.