 1. Definition of voice IC
Voice IC is converted to digital by sampling, stored in the ROM, and through the digital circuit will restore the voice signal.
ROM is Read-Only Memory , is a read only data stored in advance solid-state semiconductor memory. Its characteristics is that once the stored data can not then change or delete. Usually do not need to frequently change the data in an electronic or computer systems, data and will not disappear because the power is turned off.
2. Classification of voice IC
1) Ordinary voice IC which its playback is essentially a function of the DAC process, the ADC process data is done by computer, including voice signal sampling, compression, EQ and other processing.
- Brief explanation of the speech signal
Sampling: The analog voice signals into digital signals.
Sampling rate: the number of samples per second (byte).
Sampling bits: the number of bits in the binary condition. Generally not specified in the case, the median refers to the sound of the 8-bit sample from 00H - FFH, mute as the 80H.
- Voice IC compression technology
Because Voice information general include a large amount of data, the effective compression of voice data is necessary, it can give us in a limited ROM space entry for more audio content.
There are several ways:
Speech segmentation: the part of speech can intercept them repeatedly through the permutations and combinations to play out the full content.
Voice samples: Generally, we use the speaker frequency response curve of the IF section, less frequently used high-frequency, so the speakers sound acceptable in the case, due to lower sampling frequencies, to achieve compression, this process is irreversible and can not restored, called lossy compression.
Mathematical compression: mainly for the median sample compression, this approach is the lossy compression. For example, we often use the ADPCM compression format is to compress the voice data from 16bit to 4bit, compression ratio is 4 times. MP3 is a compressed data stream, the data related to the prediction problem, it's baud rate compression ratio of 10 times.
Digital audio referred to EQ modes, that is MP3 playback different voices, different EQ mode gives the user to listen to different voices playback, EQ mode is the most prominent personality of the individual areas, to bring more music to the user to enjoy.
2) ADC and DAC audio IC includes two processes are completed by the chip itself, including voice data collection, analysis, compression, storage, playback and other steps.
- ADC=Analog Digital Change
Analog-digital conversion (ADC) and digital / analog (D / A) conversion on the contrary, the continuous analog (such as pixel gray scale, voltage, current, etc.) through the sample into discrete digital.
- DAC= Digital Analog Change
Digital-analog converter (DAC) as the digital signal is basically the same information designed to process.
Quality hinges on the number of bits ADC and DAC. For example, Winbond W90P710 series, ADC and DAC are 32bit, close to the real sound quality. HELIOS's H224QP series together with nine of NY3 \ NY5 series, ADC and DAC are 16bit, near CD quality. SUNLINK's SLP300 series and the Alpha DAC's AMEFB series of 8bit, for the general sound quality.
|
