Method and System for the Integral and Diagnostic Assessment of Listening Speech Quality

Simple SummaryContent extracted from patent full text and abstract with AI.

This invention provides a method and system for automatically assessing the quality of speech transmitted through communication systems. Unlike traditional models, it not only offers an overall quality score, but also analyzes and diagnoses specific aspects of quality degradation such as continuity, noise, loudness, and frequency content. By comparing the input (original) and output (received) speech signals, the system evaluates unique perceptual dimensions (such as interruptions, musical tones, background and high-frequency noise, and loudness shifts), producing a detailed, diagnostic speech quality profile for use in telecommunications.

Use CasesContent extracted from patent full text and abstract with AI.

• Quality testing of speech transmission in fixed, mobile, and VoIP telecommunication networks.
• Real-time monitoring of speech quality in operational telecommunication systems to quickly detect and diagnose transmission problems.
• Design, optimization, and benchmarking of speech codecs and network components.
• Comparative analysis of network performance and device performance in handling speech signals.
• Automated quality measurement for customer experience monitoring and service level assurance in telecommunication networks.
• Diagnostics and troubleshooting of speech quality issues in multimedia and unified communications setups.

BenefitsContent extracted from patent full text and abstract with AI.

• Provides a more detailed and accurate assessment of speech quality, capturing specific degradations (e.g., noise, dropouts, unwanted tones) not available in traditional overall metrics.
• Improves diagnostic capabilities, enabling network operators to swiftly identify sources of speech quality problems and their nature (e.g., background noise, signal interruptions, frequency loss).
• Applicable to narrowband, wideband, and full-band systems, making it versatile across current and next-generation networks.
• Supports automated, objective, repeatable testing without relying on subjective human listener experiments, saving time and costs.
• Facilitates targeted optimization of network components and speech processing algorithms based on specific perceptual quality insights.
• Enables tracking of both the overall MOS (Mean Opinion Score) and underlying perceptual dimensions, aiding detailed service quality assurance.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

A method for determining a speech quality measure of an output speech signal with respect to an input speech signal, wherein the input signal passes through a signal path of a data transmission system resulting in the output signal, includes the steps of pre-processing the output signal; determining at least one of an interruption rate of the pre-processed output signal and a measure for an intensity of musical tones present in the pre-processed output signal; and determining the speech quality measure from at least one of the interruption rate and the measure for the intensity of the musical tones.