For audio equipment or theater systems, their sound cannot be measured completely using parametric indicators. The subjective evaluation of audio equipment is to evaluate the characteristics of the equipment by listening to the environment and the program under controllable conditions.
Moreover, subjective evaluation of a single device is often difficult, so standard devices or comparisons between multiple devices are usually used to obtain loudness, frequency range, tonal balance, spatial distribution, quality, and unevenness. The comprehensive effect of factors such as feelings and evaluation results.
Conditions that subjective evaluation of sound quality should meet
The evaluation of sound quality depends on many factors, such as the characteristics of the sound source (sound pressure, frequency, spectrum, etc.), the signal characteristics of the audio equipment (such as distortion, frequency response, dynamic range, signal-to-noise ratio, transient characteristics, stereo) Separation, etc.), sound field characteristics (such as direct sound, early reflections, reverberation, inter-ear correlation, reference vibration, sound absorption rate, etc.), hearing characteristics (such as loudness curve, audible range, various hearing sensations) )Wait. Therefore, this requires the evaluator to listen to various sound effects in a relatively standard environment.
(1) Selection of signal source and program material
The use of different program sources when evaluating sound quality will have a great impact on the evaluation results. Therefore, the program sources for sound quality evaluation must not only be comprehensive, including music and voice of different genres and styles, but also have strict recording and production requirements to ensure Sufficient frequency band and amplitude.
(2) the use of current high-performance system equipment
When evaluating an audio device, the product must be composed of a sound playback or pickup system by the product and other audio devices. Due to the rapid development of the audio industry, in addition to the obvious differences in objective indicators between different equipment and equipment, it also has its own "tone", so there is currently no so-called "standard sound"
This requires that the best current equipment is selected as much as possible during subjective listening evaluation, so as to reduce the introduction of additional influencing factors to subjective evaluation. Now there are many top-level "reference-level" products launched by many manufacturers in the world. Although it cannot be said that these products have reached the ideal level, their excellent performance is enough to be used as standard equipment.
(3) to meet the acoustic requirements of the listening room
The impact of the listening environment on the sound quality is unquestionable. Due to the different structures, shapes, volumes and internal surface sound absorption methods of different listening rooms, the acoustic parameters of the rooms are different. In this way, the evaluation of the same audio product in different rooms will lead to different conclusions. Therefore, for the listening room used for listening evaluation, certain acoustic conditions should be standardized to reduce the impact of the listening room on the evaluation result. In addition, the listening room should have appropriate temperature, humidity, light, to prevent these factors from affecting the mood of the audition staff.
(4) Requirements for sound quality assessors
Due to the strong personal subjective evaluation of sound quality, individual differences in listeners' occupations, physiology, emotions and cultural accomplishments directly affect the evaluation results of sound quality.
Therefore, when subjectively evaluating the sound quality of audio products, the selection of members of the audition group is very important. The members of the audition group not only have high requirements for the sensitivity of the hearing itself (referring to the conditions that the human ear itself has), but also must have an aesthetic evaluation of the sound quality. And the ability to evaluate the sound quality of musical instruments and other sound sources.
At the same time, the sound quality evaluation can also be related to the objective technical indicators of the audio product. When there is a slight difference in the sound quality of the product, what kind of sound source is heard to distinguish the sound, and what attributes are used to judge the sound quality difference.
In order to minimize the sound quality evaluation errors caused by the individual differences of the listeners, the members of an audition group may include scientific and technical workers engaged in stereo and architectural acoustics, electroacoustics, and acoustics, recording and tuning technicians, and musicians and music. Editor and film and music director, actor.
Generally, subjective evaluation of various attributes of sound quality is based on the three elements of the sound, namely changes in loudness, pitch, and timbre, such as full sound at low frequencies, full sound at high frequencies, and bright sound at low frequencies; weak sound at low frequencies, and weak high frequencies Clear voice. In the following, we will introduce several typical listening sensations based on the sound source, sound field and signal characteristics.
(1) Three-dimensional feeling and presence
The sound is not only layered, directional, but also spatial. The sound image orientation of various sound sources is accurate and clear, and there is a complete sound image group distribution. The sound distortion is small, the dynamic range is large, and the transient response is good, giving people the feeling of being immersive.
(2) Feeling of intimacy and nature
The sound changes naturally and smoothly. The loudness, tonality, reverberation, and balance are just right, without exaggeration. The momentum, style, and dynamic range conform to the natural characteristics of the sound source, making people feel kind, natural, and have pure and beautiful feelings.
(3) sense of wholeness and group
It means that the frequency is rich and coordinated, the low frequency is thick and powerful, the high frequency layer is clear, the overall momentum is magnificent without losing the delicate, the dynamic range is large and fluent and free, as if the peaks stand and give the overall beauty.
(4) a sense of integration, balance, harmony
Under the conditions of wide frequency response, low distortion, high signal-to-noise ratio, and large dynamic range, singing or band parts are relatively harmonious and balanced in terms of volume, tone uniformity, and reverberation. The entire sound field is organically integrated. Together.
In addition to the above-mentioned hearing sensations, when we subjectively evaluate the sound, we also find that the sound has the following obvious performances:
Clear-fuzzy and turbid
Means that the language has good intelligibility, and the music is well-defined. When the balance and the transient characteristics of the sound are not good, the low and reverberant sounds will bring a "blurred" and "turbid" feeling.
It refers to the coordination of the proportions of the various parts of the program, and the high, middle and low bass are properly matched. There is often an imbalance in the work itself or the band's composition, which can be balanced by tuning and pickup.
Plump-thin and dry
Refers to the fusion of sounds, especially the richness of middle and low order overtones, and the high order overtones are not too prominent. Responsive, warm, thick, and resilient. High signal levels are not necessarily full. Poor transient response, low-order overtone loss, weak mid and low frequencies, insufficient reverberation sound, and improper time delay of early reflections may all lead to a thin, dry defect
Beautiful and beautiful, shiny but not sharp. Mainly used to evaluate human voices and musical instrument sounds. Excessive treble, obvious distortion, and poor acoustic characteristics of the hall may make the sound rough.
Means that the high and full tones are full, sounds bright and active. Early reflections, reverbs, and direct sounds blend well to increase brightness. Focusing only on the sound pressure level of direct sound may not work.
The sound is slack, not sharp, comfortable and pleasant. If the pickup distance is too close, it will affect the sound diffusion and the sound will be sharp.
Keep the characteristics of the original sound source. When tuning, it will suppress the unpleasant sound, or modify the sound source in earnest, and work towards the pleasant direction, but it must not deform the characteristics of the original sound and turn the original piano into a clarinet. This is not only included in the evaluation term, but also the principle of sound reproduction.
Meaning of common listening evaluation terms
The sound is broken (splitting): severe harmonic and intermodulation distortion, there is a "snoring" sound, the flat top has been cut, and the distortion is greater than 10%.
The sound is stiff: lack of bass, too many mid-high frequencies, and the decay of harmonics at high frequencies is too short or too fast, low-frequency reverberation sounds are short, there is obvious intermodulation distortion, and transient response is not good
Explosive sound: There are too many high or intermediate frequencies, and there are two kinds of distortion.
Sound sanding: distortion in the pass band is large, there are additional higher harmonics, and transient distortion.
The sound is muddy: the transient is not good, the resonance peak of the speaker is prominent, and there are too many low or medium and low frequencies.
The sound is wide: the frequency bandwidth is good, the low and medium frequencies are good, and the reverberation is moderate.
Sound muffled: too little high or medium or high frequency, or too pointed and off-axis.
The sound is fine: the sound energy density is small, the loudness is not enough, the sound is slender and weak, the lack of low frequency, partial high frequency and high frequency, the reverberation sound is insufficient. )
There are layers: good transients, flat frequency characteristics, and moderate reverberation.
Solid sound: good mid and low frequencies, moderate reverberation, and sufficient loudness.
Sound divergence: IF is lacking, IF transients are not good, and reverberation is too much.
Narrow sound: narrow frequency characteristics, lack of high and low ends, not wide frequency band, short reverberation, and excessively prominent mid frequency.
Metal sound (lead skin sound): Individual points in the middle and high frequencies are high and the distortion is serious.
Round sound: good frequency characteristics, minimal distortion, moderate reverberation, and good transient.
Moisture: Medium and high frequencies are good, and reverberation is sufficient.
Bright sound: mid and high frequencies are sufficient, relatively flat, and moderate reverberation.
Sound spikes: too much high and medium to high frequencies.
Treble imaginary (floating): lack of mid-frequency, mid-frequency and high-frequency pointing are too sharp.
The sound is dark: lack of high and medium frequencies.
Dry sound: lack of reverberation, lack of mid and high frequencies.
Balance or Concord: good frequency characteristics and low distortion.
Roar: The loudspeaker resonance peak is severely prominent, and the transient state is not good.
Good sharpness: good mid and high frequency, low distortion, good transient, moderate reverberation
Stereo (in mono mode): flat frequency response, moderate reverberation, and good transient.
Transparency: The distortion is small, the transient response is good, the frequency response is wide and uniform, medium and high frequencies and high frequencies can be obtained, and the reverberation sound is suitable.
Presence or presence: Both frequency response and transient are good, especially mid and high frequency.
Fullness: frequency bandwidth, good mid and low frequencies, moderate reverberation.
Soft (loose): moderate and low frequency, moderate sound, loose sound, suitable loudness, slightly reverberant sound, small distortion, and bold (potential)
Good strength: good loudness, good reverb, low frequency and low frequency.