Browse Prior Art Database

A process and method for overcoming audio anomalies during recording

IP.com Disclosure Number: IPCOM000016672D
Original Publication Date: 2003-Jul-08
Included in the Prior Art Database: 2003-Jul-08
Document File: 2 page(s) / 61K

Publishing Venue

IBM

Abstract

A process for overcoming unpredictable audio anomalies in an audio subsystem. This is accomplished by separating the audio configuration into 4 parameters and configuring them independently in such a fashion as to filter out these undesired anomalies.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 2

A process and method for overcoming audio anomalies during recording

   Disclosed is a process for addressing audio anomalies that are introduced into an audio stream as a result of start up processes in an audio subsystem. One example of a source for audio anomalies in the digital audio data can be timing synchronization problems between the activation of the analog audio input signal and the Analog to Digital (A/D) Converter circuits. These anomalies can manifests themselves as spikes, pops, and/or other abnormal noise in the audio stream. Applications that expect and/or require a quality audio stream can be adversely affected if these anomalies are present. A few examples are applications that use speech recognition engines, voice recorders, transcription devices, automatic gain controllers, etc.

To address these anomalies, the audio recording configuration can be separated into 4 configurable parameters:

BufferSize - the size of a particular record audio buffer (for steady state record operation)

NumberOfBuffers - the number of record audio buffers (for steady state record operation)

PreRecordSize - the size of a particular pre-record audio buffer (to contain initial audio

anomalies present during audio start up) NumberOfPreRecordBuffers - the number of pre-record buffers (for initial hardware power

up latency operation)

In a typical embodiment the process would work as follows:

Record a sample of the audio stream.

Analyze the audio stream to determine if there are any anomalies present. If so,

Determine the position. In most cases it should be at the very beginning of the audio

   stream Programmatically configure and save the PreRecordSize/NumberOfPreRecordBuffers to...