Listener SideTone Rating

Description

Listener SideTone Rating (LSTR) is a standardized subjective testing methodology defined in 3GPP specifications such as TS 26.131 and TS 26.132 for evaluating the acoustic side-tone characteristics of speech communication terminals, including mobile phones, headsets, and landline handsets. Side-tone refers to the phenomenon where a portion of the talker's own speech signal, picked up by the terminal's microphone, is intentionally fed back at a controlled level into the talker's own receiver (earpiece or headphone). This acoustic feedback loop simulates the natural sound of one's own voice heard during normal conversation (due to bone conduction and air conduction), and its proper implementation is crucial for a natural and comfortable conversational experience. LSTR provides a quantitative measure of this perceptual quality.

The LSTR testing process involves controlled laboratory experiments with human subjects (listeners). In a typical test setup, a subject uses the terminal under test in a simulated conversational scenario, often within an acoustically insulated booth. The subject speaks predefined speech material while listening to the side-tone produced by the terminal. The subject then rates the perceived quality of the side-tone on a standardized opinion scale, such as the Absolute Category Rating (ACR) scale ranging from 1 (bad) to 5 (excellent). Key parameters evaluated include the loudness of the side-tone (it should not be too soft or too loud), its spectral balance (it should match the natural sound of the voice), and the absence of unwanted artifacts like distortion, delay, or excessive noise. The tests are repeated with multiple subjects and terminals to derive a statistically meaningful Mean Opinion Score (MOS) specifically for side-tone, which is the LSTR value.

Architecturally, LSTR is not a network function but a quality assurance metric that influences terminal design and network planning. It interacts with the terminal's acoustic design, including microphone sensitivity, receiver output, and the digital signal processing (DSP) algorithms that apply gain and filtering to the side-tone path. The role of LSTR in the broader 3GPP ecosystem is to ensure a baseline level of user experience for voice services. Poor side-tone performance can lead to users speaking too loudly or too softly, causing discomfort and fatigue, and degrading the overall conversational quality. By standardizing the measurement method, 3GPP enables manufacturers to design terminals that meet quality benchmarks and allows operators to verify terminal performance during procurement and network quality audits, ensuring consistent voice service quality across different devices.

Purpose & Motivation

LSTR was created to address a subtle but critical aspect of voice telephony quality: the user's perception of their own voice during a call. In early telephone systems, side-tone occurred naturally due to analog coupling, but its level was often uncontrolled and could be excessive (causing users to lower their voice) or insufficient (causing users to shout). As terminals became more sophisticated with digital signal processing, designers gained the ability to precisely control the side-tone path. However, without a standardized way to measure the perceptual quality of this controlled side-tone, user experience varied widely between devices, leading to dissatisfaction.

The introduction of LSTR in Release 5 provided a common, repeatable methodology to subjectively evaluate side-tone, solving the problem of inconsistent and subjective evaluations by manufacturers. It was motivated by the need to maintain high voice quality as mobile telephony became mainstream and as new form factors (like small flip phones or early Bluetooth headsets) introduced new acoustic challenges. By defining a specific testing procedure and rating scale, LSTR allowed for objective comparison of different terminal designs, driving improvements in acoustic engineering.

Furthermore, LSTR supports the broader goal of end-to-end voice quality management in 3GPP networks. While network metrics like packet loss and delay are important, the terminal's acoustic performance is the final link in the chain experienced by the user. Standardized metrics like LSTR, along with other subjective tests (e.g., for loudness, background noise), ensure that terminals do not become a bottleneck for quality. This is especially important for regulatory compliance (e.g., hearing aid compatibility) and for services like Voice over LTE (VoLTE) and Voice over NR (VoNR), where high-definition voice codecs can reveal poor acoustic design if side-tone is not properly matched to the enhanced audio bandwidth.

Key Features

• Subjective testing methodology using human listeners to rate side-tone quality on a standardized scale (e.g., MOS)
• Evaluates key perceptual attributes: loudness, spectral balance, and absence of distortion in the side-tone signal
• Defined in 3GPP specifications for terminal acoustic testing (TS 26.131, TS 26.132)
• Conducted in controlled acoustic environments (anechoic or hemi-anechoic chambers) to ensure repeatability
• Provides a quantitative metric (LSTR score) for comparing performance across different telephony terminals
• Supports quality assurance for voice services, including traditional mobile calls, VoLTE, and VoNR

Evolution Across Releases

Defining Specifications