Voice Quality vs. Sample Rate

APLUS VOICE IC allow user define different sample rate and encoding algorithm to get the best sound quality.

IC Encoding AlgorithmSample Rate
(KHz)
4bit (ADPCM) 5bit (u-Law) 8bit (PCM)
  • aP89xxx
V - V 6K ~ 22K
  • aIVRxxxx
V V V 6K ~ 20K
  • aMTPxxx
- - V 6K ~ 20K

Let's listen the sound quality base on 6KHz, 12KHz and 18KHz/8bit.

Sound quality︰

Obviously 18K Hz > 12K Hz > 6K Hz sound quality. Higher sample rate will have better sound quality. Generally speaking, suggest higher than 12KHz.

How to calculate sample rate ?

Maximum Sample Rate = Memory Size / (voice duration x Data encoding algoritm)

Ex Sound total 32sec , use aIVR8511 body
Sample Rate = 2M / ( 32sec X 4bit ) ≒ 15.6KHz (4bit ADPCM)
Sample Rate = 2M / ( 32sec X 5bit ) ≒ 12.5KHz (5bit u-Law)
Sample Rate = 2M / ( 32sec X 8bit ) ≒ 7.8KHz (8bit PCM)

Below table show how many seconds we can program for each body In 12KHz.

12KHz
4bit (ADPCM) 5bit (u-Law) 8bit (PCM)
5sec 4sec 2.5sec
10sec 8.5sec 5.5sec
20sec 16.5sec 10sec
41sec 33sec 20.5sec
83sec - 42.5sec
166sec 133sec 83sec
- - 333sec

Note: 8bit sound quality > 5bit sound quality > 4bit sound quality