What is "frequency response curve"

Decomposition of "frequency response curve": "frequency" refers to "frequency", which is the same as "tone" in sound performance; "loud" can be regarded as a speaker system (mechanical and electrical) that converts the "frequency" of the input electrical signal into Acoustic response. And this response is received by the microphone and calculated by the test instrument in the form of dB SPL logarithm. When many "frequency" response values ​​are connected together, it becomes a "curve" with peaks and valleys. Such a curve is called a frequency characteristic response curve, or simply a frequency response curve.

Speaker and frequency response curve

Is the frequency response curve of the audio system or speaker product required to be straight? Many people argue on this issue, the focus of the debate is often: good is not necessarily straight, straight is not necessarily good.

For example, the curve of a speaker near 80 Hz is more prominent, which means that this speaker is too strong for the frequency band near 80 Hz. If you play music, the bass sound will feel heavy. Or the curve of a certain speaker has a depression near 1000 Hz, which means that this speaker has weak performance in the frequency band around 1000 Hz, and the sound pressure output from the frequency band around 1000 Hz in the input signal is reduced. The sound is not the same as before.

The straightness of the frequency response curve actually tells you the difference in the gain of this speaker or sound system for sound signals of different frequency bands. The flatter the curve, the closer the gain of each frequency band of the speaker or sound system to the same. However, it is not an equal sign that the sound box or the system has the same gain and good sound for each frequency band of the input signal. why? Because the same amount of gain only expresses the same amount of amplification of the sound of each frequency band in the input signal, for example, the gain of each frequency of a system in the full audio is 30 decibels, you emit a sound of 1000 Hz, and the sound pressure level is 80 dB, the sound pressure level of the sound of 1000 Hz from the speaker is 80 + 30 = 110 dB. The sound pressure level of the 2000 Hz sound you emit is 60 dB, then the sound pressure level of the 2000 Hz sound played by the speaker is 90 dB. When not amplified by the system, the sound pressure level of your 1000 Hz and 2000 Hz sounds differs by 20 dB. Then through this system with the same amount of gain in each frequency band, the sound pressure level of the sound of 1000 Hz and the sound of 2000 Hz emitted by the speaker is also different by 20 dB, and the formation remains unchanged, ha ha. However, if your system's gain for 1000 Hz is too large (highlighted on the curve), it is not 30 dB but 40 dB, and for 2000 Hz the gain is low (the curve is recessed), not 30 dB but It's 20 decibels. Then the sound of 1000 Hz at the sound pressure level of 80 dB that you originally uttered 120 dB after passing through the system, and the sound of 2000 Hz at 60 dB that you originally uttered produced 80 dB after passing the system sound. The difference between the sound pressure level of 1000 Hz and 2000 Hz before the system is 20 dB, and the difference between the sound pressure level of 1000 Hz and 2000 Hz after the system becomes 40 dB, which is not the original difference, the team has changed shape This is also a kind of distortion. Therefore, whether the frequency response curve is straight is only representative of whether a certain speaker or a certain system performs approximately the same for each frequency band, regardless of sound quality.

As for whether it is good or not, first of all, your system should have approximately the same gain for the input signal on each frequency band (that is, the curve should be as flat as possible), so that the proportion of the sound size of each frequency band in the original signal can be amplified and restored At least, the strong should be strong, and the weak should be weak. It can truly reflect the strength of the sound, which is a good foundation. To be nice, it is more important to make a fuss about sound quality. The sound quality is bad, no matter how good the system is, it also shows a bad sound. I do n’t believe you get a Niu B speaker, use a few dozens of MP3 input to the mixer, and turn the mixer input gain to the end, playback from the Internet Download the MP3 format music, try to hear the sound. And sound quality is something internal, not just a straight curve. The curve is straight, just expressing the system's restoration of the volume. So the restoration of sound quality is estimated to be ideal. For example, it is almost impossible to completely restore the texture of the piano music recorded by the DPA microphone on the Steinway piano. This is like you listening to someone playing a violin next to you, just like you are listening to a violin song played by the same person next to the speaker. Even if you use the best sound, there will always be differences. This involves the problem of the reduction of sound quality and the reduction of the sound field, and these reduction degrees are not to say who can express it with a curve. And the quality of the sound has a lot to do with your materials, your craftsmanship, the designer's technology and artistic accomplishment. Does the master's artwork made with white jade look the same as what the street craftsman cast out with plaster?

Conversely speaking, the curve is flat, which means that the system or device restores the volume of each frequency band in the input signal to a high degree. As a sound, this is only a basic indicator, but it is also a very important indicator. For example, a sound system with good volume reduction, the input sound source signal itself is high, medium and low, etc. The volume ratio of each part is harmonious (such as the master music works recorded by the recording master, it seems that there is a fever like a UFO). Naturally, it feels harmonious. If the input signal is a song sung by a karaoke-level singer who can only scream, the high school bass is not sung originally, and it is naturally not harmonious to come out from the high-reduction sound system. However, for systems with poor reduction, for example, the frequency response curve is protruding at low frequencies, and the middle and high frequencies are a little concave. It may strengthen the bass that is not very strong, weaken the trumpet that should be strong, and play the original. Works with harmonious volume in each part may become discordant. However, if you happen to run into a situation where the player has weakened the strong bass sound, or blows up the weak trumpet sound, the negative is positive, and the original volume is not harmonious. Such a low-reduction speaker may be more harmonious than a higher-reduction speaker.

In addition, for audio products, in fact not only speakers, power amplifiers, mixers and other peripheral equipment, all have frequency response curves. According to industry standards, these devices are required to have a straight frequency response curve without adjustment. The purpose is to require these devices to maintain a faithful attitude to the volume of the signal characteristics as much as possible. If the equalizer you are using has no adjustments and the faders are flat, the frequency response curve is higher at 80 Hz and lower at 1000 Hz. Would you still want it?

Headphones and frequency response curve

It is not easy to see the quality of headphones through the frequency response curve.

The center of the earphone sound film is low frequency and the edge is high frequency.

The low-frequency end of the frequency response curve is a downward trend. In order to obtain more low-frequency kinetic energy, the spherical design in the center of the earphone is to increase his surface area to obtain bass. The frequency response curve of the middle frequency of the earphone is relatively flat because of the sound film surface Spiral pattern.

There is a large saw tooth on the frequency response curve of the high-frequency end of the headset, because there is a soft ring on the edge of the sound film to increase the elasticity of the sound film, so the resonance frequency of the soft material decreases, and the material hardens and resonates after the soft ring reaches the bonding edge. The frequency rises, forming a large saw tooth. Every headset cannot be avoided.

There are many small serrations on the frequency response curve of the high-frequency end of the earphone, and the sound film bracket and the sound film edge are bonded with burrs. It is related to the manufacturing process of the headset. If the stand and the sound membrane are integrated, there will be no such problem.

Knowing the above, we pay attention to his frequency response curve when choosing headphones. The low-end gain is greater, the high-frequency end has less sawtooth, and the intermediate frequency is flat.

Sound quality and frequency response curve

There are too many factors affecting sound quality.

First look at what sound quality is. Sound quality refers to how close the actual sound wave is to the original waveform, that is, the closer the actual sound wave played back to the waveform saved in the original audio file, the better the sound quality. Suppose there is an audio file A.wav and an ideal recording device, it can record the sound in the air without loss and save it as B.wav, then this A.wav and B.wav (from the time domain And frequency domain) The closer to the better (for more, please read the original text of chinaaudio.net main station: What is sound quality).

For a system (equipment), the amplitude-frequency response and the phase-frequency response together form the response of the entire system, and the frequency response curve generally refers to the amplitude-frequency response.

What is the process that affects the sound quality when an audio file is played from a mobile phone to being heard by someone. The general process is as follows: audio files-"Mixer" of the operating system-"DSP algorithm of the operating system (sound effects, resampling, DSP chips may be used)-" DAC-"amplifier-" headphones / speakers- "Air-" human ear.

Since air and human ears cannot be controlled, only the sound from the speakers / headphones has been studied. Almost every step before this will affect the sound quality.

The first is the mixer of the operating system, which is responsible for mixing the programs that play sounds in the system, so that each program can sound at the same time without the situation that one program will monopolize the output device and other programs cannot. Expressed in the code is to add, add the output of each program. If there is only one program playing music, that's fine, but the phone also has to deal with ringtones and reminder sounds. How does addition work? It depends on the algorithm. If it is a fixed-point addition, in order to ensure that the added value will not overflow, the two data will be shifted right and then added. The situation of floating point is more complicated, and because most of the existing audio files are in 16-bit fixed-point format, the conversion between fixed-point "-" floating points is also required, and this process will also lose precision. In short, the program will sacrifice the accuracy in exchange for dynamic range. And what if there is only one program outputting? Do n’t forget that there is also something to adjust the volume. That is to multiply each point on the waveform by a gain value. There will also be a loss of accuracy in the multiplication process. Overall, the loss of accuracy in this step of the mixer cannot be avoided. However, in addition to the input and output processes on the mobile phone, floating point operations are performed in the middle, and the loss of accuracy generally does not exceed -90dB, which is generally inaudible.

Then comes the DSP algorithm part. The sound effect (bass enhancement, increased sense of space, etc.) is subjective and does not belong to the category of "sound quality", so it will not be discussed. Assuming all sound effects are turned off, the only thing left is resampling. For mobile phones, resampling exists because a DSP chip often supports only one output sampling rate, or a DAC only supports one input sampling rate, and in most cases this sampling rate is 48kHz. This is because if you want to support different sampling rates, especially non-integer sampling rates like 44.1kHz and 48kHz, you need to equip crystals with different frequencies. For various reasons, it is easier for the crystal to generate a clock frequency of 48kHz. However, due to various historical reasons, most of the current music is 44.1kHz, so it will undergo a 44.1kHz-》 48kHz resampling. Non-integer multiple resampling will greatly lose accuracy. Don't think that the sound quality will be better if the sampling rate becomes higher. The best sound quality is output directly without re-sampling. The effect of resampling on sound quality depends on the resampling algorithm, and poor quality algorithms can cause severe distortion.

Next is the DAC, the digital-to-analog converter. This is a module that has a significant impact on sound quality. The frequency response of the DAC is also easy to be straight, but there are many other parameters that need to be referenced to measure the sound quality of the DAC. The quality of the DAC basically depends on the manufacturer and model of the chip itself, so there is nothing to say. Good equipment will use more high-end DAC.

Then there is the amplifier. Relatively speaking, this part is relatively easy to achieve a straight amplitude-frequency curve. But the phase frequency is not necessarily. (At present, the frequency response of the amplifier is easy to be straight)

Finally, there are headphones / speakers. Generally speaking, their amplitude-frequency curve is difficult to be straight, which is largely because the height of the frequency that the sound-emitting unit can emit is inversely proportional to its size. So don't expect earbuds to emit effective low frequencies at all. This is also the main reason why headphones generally have better sound quality than earplugs or hanging ears. For speakers, two-way frequency, three-way frequency, or even multi-frequency are often used, that is, multiple sound-emitting units are responsible for different frequency bands, among which there are problems such as filtering and processing band connection. From the perspective of the entire audio stream, headphones / speakers are the most influential parts of sound quality. All the music you put in your phone is lossless music. The phone supports direct output of 44.1kHz. The DAC uses the best chip. The amplifier has almost no distortion. As a result, you use a pair of headphones bought on a street stall for 50 yuan. The sound quality It's just a scum.

Overall:

1. Can the frequency response curve reflect the sound quality?

can. In theory, the flatter the frequency response curve is, the better the system response will be. But looking at a frequency response curve is very incomplete.

2. To what extent does the frequency response curve of the amplifier determine the sound quality?

Very little.

3. For mobile phones, what are the parameters that affect sound quality worthy of attention?

The mixer and resampling algorithm are the same or similar for all mobile phones.

The amplifier is more important. At present, the amplifier of the mobile phone can already achieve a good system response, so everyone is not much different.

DAC is more important, depending on the chip model.

The decisive part is still in your playback device. Using a better headset or speaker is more effective than anything.

Other commonly used parameters for evaluating sound quality include distortion and signal-to-noise ratio.

Tone and frequency response curve

Let's start with the analysis of "frequency": we will find the same "tone" (frequency) sound in different musical instruments, but its "timbre" is different, so what factor determines the tone of the instrument?

Answer: It is because the harmonic components contained in the "tone" (frequency) are different.

We know that sound is produced by vibration. It is almost impossible for an object to vibrate back and forth in accordance with a certain period. That is to say, when an object sounds, it will also emit many waves (harmonics) of different frequencies. Because of the small phase difference between the waves of many different frequencies (the time between the waves is very short), people cannot distinguish them separately, so these waves will be mixed together to give a whole sound experience, and this feeling is called timbre.

Some people questioned that although the "tone" is the same in actual musical instruments, it is difficult to ensure that the contrasted sound pressure / loudness can be consistent when blowing, pulling, and playing, so the sound we hear will certainly be different.

In order to exclude this point of view, an experiment can be done: in theory, when two sounds with the same sound pressure level are superimposed, the total sound pressure level on the reference axis will increase by 3dB. We took two loudspeaker units with the same sound pressure level at the same frequency to superimpose the sound, and then compared the listening with a single + 3dB unit. The end result is that the sounds of sounds with the same sound pressure level still feel very different. (At this time, only when the following conditions are met: that is, the harmonic component of the superimposed sound pressure is the same as the harmonic component of a single + 3dB speaker, the timbre will not be easy to distinguish)

Since every sound in the instrument contains sound waves of many frequencies, how do we distinguish the pitch (frequency)?

Answer: The frequency with the largest relative amount of a certain frequency in a sound determines the pitch of the sound. For example, if a sound contains 3 units of 444Hz (la sound) and 1 unit of 222Hz frequency, then what we hear is la sound. And there are 3 units of 444Hz, 1 unit of 333Hz, then we still sound la sound, but the tone is different.

After explaining the relationship between "frequency", "loud" and "tone", let's explain some problems that should be paid attention to in the test curve during speaker development Tell me about the problem of the difference in sound pressure level in the mid-band).

1. In the loudspeaker specifications of many companies, the average sensitivity column will be marked as follows, such as: 82dB ± 3dB. Therefore, many people think that they have completed the development task when copying the speaker within the allowable range of 82dB, and the sample is judged to be NG after being heard.

Because the relationship between dB SPL values ​​is logarithmic, this shows that within 1 dB there is also a relatively wide range of sound pressure (strong). The human ear can distinguish the difference of 1dB sound pressure level in the more sensitive frequency band. Therefore, in the R & D process, the dB SPL of the mid-band should be controlled within ± 0.5dB as much as possible.

2. Some engineers made the SPL value difference within 1dB, or even closer. At this time, the sound still feels different. At this time, the test environment and the way of curve performance should be considered. When the test environment is poor, the environment interferes greatly with the test microphone, and several curves tested on the same speaker will have large errors. Therefore, repeat the test and exclude the environmental impact factor before analyzing.

The test environment is better. For example, in a standard soundless room, there is no human operation error, and the curve is controlled within 1dB. At this time, it is necessary to consider the number of different points of the curve and the smoothing mode. The difference between the number of points and the smoothing mode on the details of the frequency response curve is very different. (In the test frequency range, the number of response values ​​of the instrument to different frequency points is selected. The more points, the more accurate the test. The sampling and smoothing method of the response amplitude of the frequency point in the frequency range, such as 1 / 6oct, 1 / 12oct, 1 / 24oct, 1 / 48oct, etc., the larger the denominator, the more accurate the data). Therefore, if you want to understand the changes in these details on the frequency response curve, you have to pay: a good standard test environment.

It should be noted that when using multi-points and without smoothing curve comparison, in addition to environmental factors, some problems will be encountered. For example, it is mentioned in the above harmonic: because of the back-and-forth vibration of an object, it is almost impossible to vibrate at a certain period all the time. At this time, there will be a phenomenon of sound pressure level drift at certain frequency points. At this time, according to the function of the instrument, the frequency range of the original test can be segmented, more points can be scanned, and the response trend of the frequency band can be understood in multiple directions.

3. Eliminate the test error factor, compare the harmonic components of speakers with very close frequency response curves, and find out the direction of the next improvement. General instruments for testing harmonic distortion can easily compare harmonic components.

Conclusion:

When the speaker frequency response curve is the same, its timbre is determined by its harmonic content. When the speaker frequency response curve is the same, if the tone is different, you need to compare the speaker harmonic curve. During the testing process, it is necessary to pay full attention to the impact of operators, test instruments, usage methods, test environment and treatment tools on the test results, and use as many points as possible to compare with less smooth methods.

Universal Back Sticker

Universal Back Sticker, Back Film, TPU Back Sticker, Back Skin Sticker, PVC Back Sticker, Back Skin,Custom Phone Sticke,Custom Phone Skin,Phone Back Sticker

Shenzhen Jianjiantong Technology Co., Ltd. , https://www.tpuscreenprotector.com