At the recent AES convention in Munich, audio technology house APTX (formerly APT, Audio Processing Technology) announced that it is developing a new, scalable variant of its “apt-X” audio compression algorithm. The new codec is called “apt-X Scalable,” and is intended to be used as a single-codec solution in products that process a range of audio inputs, such as mobile devices that include voice, music, ringtones, and other forms of audio.
According to APTX, apt-X Scalable will dynamically adapt its encoding scheme to incoming audio characteristics, a capability APTX refers to as “cognitive” encoding. According to APTX, the codec will select an encoding scheme based on the characteristics of the incoming audio signal, the limitations of the system hardware and network, and possibly some higher-level inputs about performance priorities; for example, whether to prioritize bit rate or coding delay. Apt-X Scalable is expected to be available for license in the fourth quarter of this year.
The announcement of apt-X Scalable follows the company’s announcement of “apt-X Lossless” last month; apt-X Lossless is intended for use in high-definition digital audio applications. (Fraunhofer also recently introduced a lossless audio codec, HD-AAC.)
APTX was founded in 1989 and is based in Belfast, Northern Ireland. The first apt-X codec was developed in the early 1990’s and was used in codec chips for broadcast and cinema applications; later, the company began offering its codec algorithms for license. APTX codecs have been widely used in broadcast applications in Europe, though they are not particularly well known in the U.S.
In addition to the two new codecs announced this year, APTX offers several other versions of apt-X in an effort to address the needs of a wide range of audio applications. According to APTX, the fundamental advantage of the apt-X codecs is their low latency and high audio quality; apt-X codecs have varying bit-rate and audio quality characteristics with delays of as little as 1.9 ms. Unlike many popular audio codecs, the apt-X codecs use ADPCM coding rather than perceptual coding techniques. The codec family is shown in Table 1.
|
Codec Variant |
Characteristics |
Sample Applications |
|
Standard apt-X (early 1990’s) |
Low latency (1.9 ms delay, 4:1 compression) |
Initial codec developed to provide low-latency, high quality compression |
|
Enhanced apt-X (2000) |
Better audio quality than Standard apt-X |
Pro audio, broadcast audio |
|
apt-X Live (2006) |
Low bandwidth (1.9 ms delay; 8:1 compression) |
Digital wireless microphones and other applications with low-bandwidth, low-delay requirements |
|
apt-X Lossless (2009) |
Lossless audio encoding of up to 24-bit resolution at 96 kHz. Can also operate in a hybrid lossy mode that limits the peak bandwidth |
HD audio, multimedia, music storage |
|
apt-X Scalable (announced 2009) |
Automatically scalable based on incoming audio characteristics and pre-set optimization priorities |
Devices that require multiple codec performance levels (e.g. mobile devices with voice, music, ringtones, etc.) |
Table 1. Apt-X audio codecs.
In April the company announced that its broadcast hardware division had been sold to the Audemat Group, leaving APTX free to focus on developing and licensing the apt-X codecs. This strategy is the reverse of that taken by some other algorithm providers, who have started with an algorithm and then moved into the hardware business in an effort to provide more complete, easy-to-integrate solutions and attract more users for the algorithms. (Recent examples include Samplify Systems and Audience, Inc.). APTX decided to sell its hardware division in part to clarify and refocus its licensing business; in some cases the company found itself trying to license its algorithms to hardware companies that were competitors to its own broadcast hardware business. Selling the hardware business eliminated that conflict, and APTX has said that it will use the proceeds to expand its licensing business.
Add new comment