Audio Jitter Buffer Calibration for Low-Latency Output

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Audio data streams transmitted over jittering paths experience varying latency, leading to increased output latency due to the uncertainty of the first audio packet's arrival time, which can result in unnecessary delays in audio output.

Innovation Solution

A method is implemented where calibration packets are used to determine the distribution of latencies, creating a reference time grid and an output time grid that minimizes latency by shifting the output time grid relative to the reference grid based on the offset of the fastest calibration packet, allowing for the earliest possible output of audio packets.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a jitter buffer is used to handle varying transmission latencies, then the reliability of audio output is improved, but the latency of audio output increases

Engineering Contradiction:
Improveaudio output continuityVSAvoidaudio output latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs a calibration phase before actual audio transmission to measure the transmission path characteristics. By preliminarily determining the jitter window length and offset values through calibration packets, the system can configure the jitter buffer optimally in advance, reducing the latency added during normal operation while ensuring reliable audio output continuity.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent dynamically adjusts the jitter buffer parameters based on measured transmission characteristics. Instead of using a fixed, conservative jitter buffer size, the system adapts the buffer length and offset values according to the actual jitter window measured during calibration, optimizing the balance between reliability and latency for the specific transmission path.

Inventive Principle:
Principle #15Dynamics

2Reliability

If the first audio packet is assumed to be an early packet and delayed by the full jitter buffer length, then audio output continuity is ensured, but unnecessary latency is introduced

Engineering Contradiction:
Improveaudio output continuityVSAvoidaudio output latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary calibration to measure the actual offset of the first packet relative to the jitter window. By determining the true position of early packets through calibration packets and offset calculations, the system avoids the conservative assumption of delaying all packets by the full jitter buffer length, thereby reducing unnecessary latency while maintaining output continuity.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the timing parameters for packet output based on measured transmission characteristics. Instead of using a fixed delay equal to the jitter buffer length, the system adjusts the output timing offset according to the measured offset values from calibration, optimizing the balance between ensuring continuity and minimizing latency.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10686897B2Method and system for transmission and low-latency real-time output and/or processing of an audio data stream
Publication Date: 2020.06.16 SENNHEISER ELECTRONICS GMBH & CO KG
  • US10686897B2 patent drawing
  • US10686897B2 patent drawing
  • US10686897B2 patent drawing

AI summary

A method for transmission and low-latency real-time output and/or processing of an audio data stream that is transmitted from at least one transmitter to at least one receiver over a jittering transmission path. The method includes a calibration for determining a distribution of latencies in transmission of packets of the audio data stream, whereby a group of packets of the audio data stream is used as calibration packets and wherein a reference time grid and an offset of a fastest calibration packet are determined. Then, a shift of an output time grid for audio output and/or processing, based on the reference time grid and the determined offset of the fastest calibration packet, and the audio packets of the audio data stream are provided according to the output time grid for audio output and/or processing.