Question 1

What is the STIPA test signal and how does it work?

Accepted Answer

The STIPA (Speech Transmission Index for Public Address) test signal is defined in IEC 60268-16 Table F.1. It consists of pink noise simultaneously modulated at 14 different frequencies across 7 octave bands (125 Hz to 8 kHz). Each octave band carries two modulation frequencies, selected to avoid overlap and allow simultaneous measurement. The modulation frequencies range from 0.63 Hz to 12.5 Hz, representing the temporal patterns present in speech. When this signal passes through a sound system and room, the modulation depth is reduced by reverberation, noise, and distortion. By measuring the reduction in modulation depth (the Modulation Transfer Function) at each frequency, the STI can be calculated in a single 15-second measurement. SonaVyx generates the STIPA signal and analyzes the captured response to extract the MTF and compute STI.

Question 2

How long does a STIPA measurement take?

Accepted Answer

The minimum STIPA measurement time is approximately 15 seconds per position. This is determined by the lowest modulation frequency in the test signal (0.63 Hz), which has a period of 1.59 seconds — at least 10 cycles are needed for reliable measurement, giving about 16 seconds. SonaVyx Quick mode uses a single 15-second capture. Pro mode takes three consecutive measurements and averages them, totaling about 50 seconds per position (including gaps between measurements). For a complete venue assessment, measure at 6-12 positions covering the audience area. At 3 positions per speaker zone, a 4-zone venue requires 12 positions, taking about 10-20 minutes total. SonaVyx measurement wizard guides you through the positions and tracks which areas have been covered.

Question 3

What equipment do I need to measure STI?

Accepted Answer

You need: a source device to play the STIPA test signal (SonaVyx can generate it through any connected speaker or the venue sound system), and a measurement microphone to capture the response. Feed the STIPA signal into the sound system at the normal speech input level (equivalent to someone speaking into the microphone at a normal distance). Place the measurement microphone at ear height in the audience. For accurate results, use an omnidirectional measurement microphone. A phone microphone works for screening measurements but may not meet the accuracy requirements of formal compliance testing. The STIPA signal must be played at a level representative of normal system operation — too loud or too quiet will give incorrect results because the signal-to-noise ratio at the listener affects the STI.

Question 4

How do I measure STI from an impulse response instead of a STIPA signal?

Accepted Answer

STI can be calculated directly from a room impulse response using the indirect method described in IEC 60268-16 Annex A. The Modulation Transfer Function is computed from the squared IR envelope in each octave band. This method yields the STI contribution from the room acoustics (reverberation) but does not capture the effects of system distortion or fluctuating ambient noise. To include ambient noise, measure the noise spectrum separately and apply the noise correction formula from the standard. The IR-based method is convenient because a single sweep measurement gives both RT60 and STI. SonaVyx computes STI from impulse responses automatically and provides the option to enter measured ambient noise levels for correction. The result is displayed alongside the STIPA measurement result if both are available.

Question 5

What is the correct level to play the STIPA test signal?

Accepted Answer

Play the STIPA signal at the normal operating level of the sound system for speech. If the system is used for paging, set the level to the typical paging volume. If used for speech reinforcement, set it to the level a speaker would produce through the system. The signal-to-noise ratio at the listener position critically affects STI: too quiet (low SNR) will underestimate the STI, while too loud may introduce distortion that artificially reduces STI. A good practice is to set the level so that the LAeq at the listener position is 60-75 dBA, which is the typical speech reinforcement range. If the ambient noise is high (e.g., 65 dBA in a factory), the system must produce correspondingly higher levels to maintain intelligibility. SonaVyx displays the received signal level during measurement.

Question 6

Can I measure STI during a live event?

Accepted Answer

STI requires the standardized STIPA test signal — you cannot derive it from arbitrary speech or music. However, you can measure during breaks, sound checks, or before/after the event when the system is available. An alternative is to measure the impulse response during the event (using the system playing music as the excitation, as in a transfer function measurement) and derive STI from the IR. This indirect method captures the room acoustics but not the real-time noise conditions. For ongoing monitoring, some venues install permanent STIPA generators and microphones. SonaVyx offers both direct STIPA measurement (requires control of the signal chain) and IR-derived STI (can be obtained during normal operation).

Question 7

How many measurement positions do I need for an STI survey?

Accepted Answer

The number of positions depends on the standard being applied and the venue size. IEC 60268-16 recommends at least one position per distinct coverage zone. For voice alarm compliance (EN 54-24, BS 5839-8), measure at a grid of positions covering the entire area, typically 2-4 m spacing. For general assessment: 3-6 positions per speaker zone, including on-axis (best case), off-axis (mid coverage), and edge of coverage (worst case). Report the minimum STI found, as this represents the worst-case intelligibility. Total positions for a medium venue (500 m²) might be 12-20. For large venues, prioritize areas far from speakers, under balconies, and at the edges of coverage. SonaVyx records each position result and generates a summary showing the spatial distribution of STI scores.

Question 8

What environmental conditions affect STI measurement accuracy?

Accepted Answer

Several conditions affect measurement accuracy. Ambient noise reduces the modulation depth at the microphone, lowering the measured STI. This is a real effect (noise degrades intelligibility) and should be captured if you want to assess the actual conditions. Temperature and humidity slightly affect air absorption, which reduces high-frequency modulation transfer at long distances. Wind noise on an outdoor microphone contaminates the measurement. Time-varying noise (intermittent machinery, passing traffic) creates variability between measurements. For reproducible results: measure during consistent noise conditions, note the ambient LAeq, use a windscreen outdoors, and take multiple measurements to check consistency. SonaVyx records ambient noise level during each STIPA measurement for reference and noise correction.

STI Measurement Guide: STIPA Test Signal, Procedure, and Equipment Requirements

Frequently Asked Questions