With the progress of AI technology, audio products are constantly evolving, and the functions of intelligent audio devices are becoming more and more powerful. Intelligent audio enhancement technology refers to the use of computer science and artificial intelligence technology to deal with audio signals to improve their sound quality, volume and sound effect. Audio enhancement technology is very important in the audio field, mainly including volume gain, which can enhance the loudness of sound, dynamic range compression, sound balance, surround sound processing, echo and noise elimination, noise suppression, audio balance and so on by increasing the volume of audio signals. Realizing these mainly depends on AI module and audio processing module. This issue mainly explains the application on intelligent audio module.

As a component integrating multiple audio processing functions, it is usually used to realize the input, output, processing and control of audio signals in electronic equipment. Core components include microphone array, digital signal processor, audio codec, discrete devices, network interface, audio input module, audio output module, processing module, control module, communication module, expansion interface and so on. This component plays a key role in receiving and processing transmitted audio signals. Intelligent audio modules are widely used in audio equipment, headphones, medical electronics, mobile phones, televisions, car stereos, cinema equipment, DVD players, instruments and industrial control products.

Discrete devices play the functions of small signal amplification and voltage regulation, rectification and filtering, and circuit protection on the intelligent audio module. The specific circuit applications include rectification and filtering circuit, small signal amplification circuit, charging management and control circuit, step-down circuit, switching circuit and protection circuit.

Take the echo cancellation circuit as an example, this part will use diode products. The challenge to the accuracy of speech recognition for smart speakers is that the microphone will pick up the ambient noise. In order to filter the noise, one technology is echo cancellation, which can identify the digital signal of the music being played and then remove the signal from the microphone input. In order to filter out the environmental noise better, the audio amplifier with echo cancellation function has an SDOUT signal, and then the SDOUT signal is fed into the SoC to filter out the music signal received by the microphone.

Take the audio output module as an example. The sound source module of smart speakers is usually divided into two parts, one is the digital part responsible for outputting digital signals, and the other is the decoding part responsible for digital-to-analog conversion. The power amplifier is also divided into pre-amplifier and post-amplifier. The pre-amplifier is mainly responsible for signal voltage amplification, volume control, multi-channel sound source switching, tone adjustment and impedance matching, while the post-amplifier is specifically responsible for power amplification, and both parts will use diode products.

The purpose of power amplifier is to provide enough current driving capacity for the load to realize power amplification. Class D power amplifier is also called digital amplifier, which uses the switching circuit of extremely high frequency to amplify audio signals, which has the advantages of high efficiency and small size. A typical class D power amplifier topology consists of triangular wave generator, comparator, power output stage and LC low-pass filter. In the full-duplex voice interaction system of smart speaker, the quality of power amplifier is very important, because AEC echo cancellation is very sensitive to signal distortion, and the performance of diode has a great influence on power amplifier.

As a manufacturer specializing in discrete devices, Heketai products mainly provide products and services such as bridge stacks, diodes and triodes for intelligent audio modules. The best-selling diode models include 1N4148, 1N4148, M1, etc., and the triode models include S8050, S8550, MMBT3904, MMBT3906, etc. The product is stable and powerful, and it is a good partner for audio customers.