TOP 5: Bluetooth audio codecs: aptX + (HD), LDAC, AAC and SBC
There are 5 main Bluetooth audio codecs, with which sound is transmitted from a source to headphones (or speakers) via Bluetooth: SBC, AAC, aptX, aptX HD, and LDAC. As wireless connectivity becomes more and more popular, it is a good idea to understand what Bluetooth codecs are and how they work.
Bluetooth aptx
It is worth noting 2 points at once:
-Bluetooth sound is still noticeably worse than wired sound in all quality parameters;
-At this stage, wireless sound with high-quality codecs can meet the needs of most users.
We will no longer focus on these “axioms”. In general, they are clear. And many people with experience in listening and comparison are unlikely to argue with them.
How to choose, change and learn the Bluetooth codec on an Android phone?
Which is better: listening to music via cable or via Bluetooth?
Basics of Bluetooth audio streaming
Bitrate
What is bitrate, I think everyone knows. This is the number of bits that are used to transfer data per second. The higher it is, the better, obviously. But the more resources required, in particular power consumption, disk space, etc.
Compression has long been used to optimize resources, lossy and lossless. Lossless compression uses algorithms to reduce the bit rate and file size without altering the original. That is, without loss, you can restore an exact copy of the original uncompressed file.
Maximum bit rate of bluetooth codecs
Maximum bit rate of Bluetooth codecs
Lossless processing allows you to reduce the bit rate a maximum of 2 times. More or less. For example, for a CD quality file (1411 kb / s) compressed without loss, we will see values of 770-900 kb / s. Anything below is lossy compression.
When the bit rate and / or file size needs to be further reduced, it is compressed by discarding the “irrelevant” bits of information. That is, the selected codec, according to its algorithms, “decides” in which frequency range there is no data (or it can be neglected), and discards it. The uncompressed original cannot be recovered from the resulting file.
The algorithms for the codecs are different: some exclude only high-frequency sounds, some are based on psychoacoustics. For most, the difference between a good lossy track and its lossless version is negligible or non-existent.
In addition to “discarding” in lossy compression, “overlapping” information blocks are added. This part of the algorithm aims to flatten the signal and correct for quantization errors.
Sample rate and bit depth
The sample rate shows how many “discrete” segments the analog signal is divided into for digital recording. The larger it is, the closer the “figure” is to the original. Measured in Hertz. This concept is inextricably linked to the frequency range.
A person hears sounds from 20 Hz to 20 kHz. It has been shown that to quantize (digitize) a lossless audio signal, the sample rate must be more than 2 times the frequency range of the original audio signal. Consequently, the 44.1 kHz CD quality is sufficient: the creators of the format relied on this law.
Bit depth determines how many bits are needed to quantize each of the preceding segments. The bigger the better. It is logical that the audio format (or codec) is considered better the higher its bit rate and sample rate. 16 bit 44.1 kHz for CD versus 24 bit 88.2 (or more) kHz for Hi-Res Audio.
Sampling rate
Sampling rate
Background noise level
The background noise level determines when the desired signal becomes indistinguishable from the “noise” of the codec compression algorithm, in this case. The lower this level, the wider the dynamic range of the final signal, the quieter the sounds we can distinguish.
Noise level LDAC (990 kbps) (CD)
Example: LDAC noise level 990 kbps
On the charts, it is more correct to navigate from the highest peak, it can drown out the lowest sounds. It is not only the level of the peaks that is important, but also how smooth (“dense”) they are. Typical background noise levels are indicated in graphics by a dotted line for 16-bit quality and a top red area for 24-bit quality.
All codecs “try” to save bandwidth. In addition to all other settings, quantization noise is removed, if possible, in the HF range above 20 kHz. A person does not hear these sounds, so the algorithm is justified.
