A new generation of audio decoders

Edited by the Electronicstalk editorial team Feb 28, 2001

Flint has expanded its range of audio decoders with a family of single-chip solutions that provide longer playing times in smaller packages with lower power consumption.

Flint has expanded its range of audio decoders with a family of single-chip solutions that provide longer playing times in smaller packages with lower power consumption.

The Micronas MAS35x9F series encompasses three industry standards: MPEG layer 2/3 and MPEG2-AAC audio stereo decoding, together with the G.729 Annex A speech compression and decompression.

Additional functionality such as CELP voice decoding and Micronas SC4 (ADPCM) encoding/decoding is possible using downloadable software.

There are six devices in the family, from the MAS3509F, which incorporates all three standards, through two dual standard versions, to compact single-standard devices.

All are based on RISC architecture and DSP technology, enabling efficient processing of audio data.

The MAS 35x9F decoding block accepts compressed digital data streams as serial bitstreams (or parallel format) and provides serial PCM or S/PDIF output of decompressed audio.

In addition to signal processing functions, the IC incorporates a high-performance stereo DAC, headphone amplifiers, a stereo ADC, a microphone amplifier, and two DC/DC convertors.

This all-in-one solution is ideally suited for highly optimised memory-based portable music players with integrated speech decoding function.

Despite such a high level of integration, these ICs require very little power.

Their built-in DC/DC convertors allow the chip to run on 2V simultaneously with other components in the system that require higher voltage, like Flash memory.

System manufacturers can produce cheaper, smaller, and more attractive looking MP3 devices that require only a single battery.

The MAS3509F, with the highest degree of integration, consumes only 65mW whereas previous decoders, as a result of optimising the design of building blocks to 2.0V.

Besides MP3, the new decoders also support AAC (Advanced Audio Coding) and various formats for voice recording and has designed-in functions to comply with the Secure Digital Music Initiative (SDMI) requirements regarding the security of copyrights and reproduction rights.

AAC is also defined within MPEG 2, providing compression rates up to 16:1.

This IC decodes the 'low complexity profile' that is especially optimised for portable applications.

The MAS 35x9F also implements a voice encoder/decoder compliant with the International Telecommunications Union's standard G.729 Annex A technology, which is ideal for use in memory-based or broadcast applications.

Furthermore SC4 - a proprietary Micronas speech codec technology - can be downloaded to the MAS 35x9F to allow recording and playing back speech at various sampling rates.

Additionally, the chips feature Micronas' Perfect Audio (MPA), integrated audio algorithms that guarantee the highest sound quality including automatic volume control, a five-band or parametric equalizer, bass boost, and dynamic bass enhancement or 3D sound.

MPA is available to system developers at no extra license fees.

Processing for these standards is delivered from a core developed specifically for playing psychoacoustic compressed data flows such as MP3 or AAC.

The on-chip ADC with automatic gain control can convert analogue signals for speech recording into higher-quality digital signals, whilst the D/A converter reaches a signal-to-noise ratio of 95dB and delivers an output of 5 to 15mW.

Flint has all six variants available, delivered in 1.4 mm high, 64-pin TQFPs.