How to use AP525 to test the audio performance of Telink hardware (Part 1) - Introduction to basic audio parameters

Sound exists in every corner of life and is an important way for people to understand the world. In order to understand sound more concretely, people use various terms to describe it, such as frequency response, sampling frequency, bit depth, sound pressure, THD+N, etc. Through audio test instruments, sound can be expressed through specific data or curves, helping people to more reasonably evaluate whether an audio system meets the requirements or how to optimize it to meet the requirements. Audio Precision (AP) is a test tool currently trusted by many audio practitioners. In the future, audio-related tests will be introduced based on AP525 and Telink TLSR9518A EVB.

This article mainly sorts out the audio parameters frequently encountered in the test:

Sample Rate

Bite Depth

Frequency Response

Average level (RMS Level)

Gain

Noise Level

Signal-to-Noise Ratio (SNR)

Total Harmonic Distortion + Noise (THD+N)

Delay

1

Sampling rate

ADC sampling rate: The number of times an analog signal can be collected and converted per second, which determines the quality and naturalness of sound restoration. The higher the sampling rate, the more sound details can be collected and retained, and the more realistic and natural the sound can be restored later. Since the human ear's hearing range is approximately 20Hz ~ 20kHz, theoretically speaking, a sampling rate greater than 40kHz can cover the audio range that the human ear can hear. The 44.1kHz and 48kHz sampling audios commonly seen in life are all lossless audio.

DAC sampling rate: The number of signals extracted from the signal per second determines the maximum frequency component that can be output.

2

Bit Depth

The bit depth of the audio determines the dynamic range that can be expressed. For example, the dynamic range of a 16-bit ADC = 20*lg(2^16/1)≈96dB, and the dynamic range of a 24-bit ADC = 20*lg(2^24/1)=144dB. Therefore, the greater the bit depth, the greater the dynamic range, which can record more details about the level amplitude, and the final restored sound is more realistic.

The sampling amplitude of the red dots in the figure below can be distinguished at a bit depth of 16, but cannot be distinguished at a bit depth of 12.

Figure 1 Sampling at different bit depths

3

Frequency Response

For audio equipment, frequency response can be understood as the processing or reproduction of different frequency signals by the audio equipment. For audio signal frequency, the range of 20Hz~20kHz is generally concerned. Ideally, different frequency signals with the same input amplitude will have the same output amplitude after passing through the audio equipment, that is, they tend to be a straight line. In actual applications, the frequency response of audio devices is not flat, or the frequency response curve is adjusted into a "curve" of different styles according to the required effect.

Figure 2 Ideal frequency response curve

Figure 3 Adjusted frequency response curve

4

Average level

The signal amplitude of a digital or analog signal after it passes through an audio system.

5

Gain

The degree to which the analog or digital signal is amplified or attenuated by the system after it passes through the audio system. Usually for the ADC and DAC paths, there will be corresponding analog gain and digital gain that can be adjusted.

Figure 4 ADC path gain

Figure 5 DAC path gain

6

Noise Floor

The background noise intensity output by the audio system is the total noise in the audio system except the useful signal.

When the background noise is large, you can hear obvious "rustling" noise or "snowflake" sound;

When other interference signals such as power supply or radio frequency introduce a specific frequency signal, noise of the corresponding frequency will be heard

Figure 6: Noise floor frequency domain diagram

7

Signal-to-Noise Ratio

The ratio of the useful signal strength to the noise signal strength in the audio signal. The unit is dB, and it is an important parameter used to measure the quality of audio. If the signal-to-noise ratio is too low, the noise may interfere with the audio signal, and the actual experience will feel that the sound quality is fuzzy, and the clarity of music or call voices is reduced. Therefore, if you pursue high sound quality and high clarity, then the signal-to-noise ratio must be an important indicator to refer to.

Figure 7 Main frequency signal and background noise

8

Harmonic distortion + noise

Total harmonic distortion (THD) refers to the percentage of the energy of all harmonics of a signal within the measured frequency range to the energy of the main frequency signal, and is used to describe the strength of the signal harmonics.

THD+N is the addition of background noise on the basis of THD. It can be used to evaluate the strength of harmonics and background noise signals. Harmonics and background noise are both "unhelpful" signals for audio systems. As long as one of them is relatively poor, it can be monitored from THD+N.

Figure 8 Schematic diagram of THD+N poor signal frequency domain

9

Delay

The time required for an audio signal to be input into an audio system and then output from the audio system. In actual audio products, the smaller the delay, the better the experience for applications such as games and videos that are sensitive to delays.

Figure 9 Audio system delay diagram