Paris -- Dolby Laboratories may no longer have a lock on the multichannel-audio market, thanks to the emergence of a new surround-sound technology that stole the limelight at the Audio Engineering Society (AES) Expo here last week. MPEG Surround allows broadcasters or content owners to add high-quality surround sound, with very little overhead in bandwidth, to MP3 files or other MPEG audio streams for Internet Protocol TV, mobile entertainment and Internet streaming applications, proponents said.
MPEG Surround, which is scheduled to become a Final Draft International Standard in July, is a very efficient parametric coding technology. It essentially reduces bit-rate requirements for high-quality multichannel audio compression while maintaining backward compatibility with exist- ing stereo equipment.
Now proponents say it's up to chip makers to step in with the devices needed to bring the new audio technology to market.
While details on intellectual-property rights for MPEG Surround are still being worked at, "they will be made available in a fair, reasonable and nondiscriminatory manner," said Bernhard Grill, head of the audio department at the Fraunhofer Institute for Integrated Circuits IIS.
The technology can transform mono or stereo audio signals compressed in MP3, MPEG-4 Advanced Audio Codec or MPEG-4 HE-AAC into high-quality multichannel surround audio "with only a negligible increase in data rate," Grill said.
The MPEG Audio standardization group developed the system by combining two submissions: one from Fraunhofer IIS and partner Agere Systems Inc., the other from partners Coding Technologies and Philips. The group said these two submissions outperformed others while complementing each other's performance in cer- tain areas. The public demonstration last week at AES marked the first joint presentation by all four parties.
Could MPEG Surround become a Dolby killer? Perhaps, its backers say--especially in applications that must operate under severe transmission-channel bandwidth limitations. In fact, MPEG Surround marks a sharp departure from traditional 5.1-channel Dolby Digital, which requires proprietary multichannel audio compression separate from MP3 and other MPEG audio codecs. Transmitting high-quality multichannel audio signals like Dolby Digital also eats up "roughly 2.5 times [the] bandwidth," said Fraunhofer's Grill.
If broadcasters are required to "simulcast"--broadcast multichannel audio signals simultaneously with existing stereo audio channels--bandwidth demands grow even greater. Further, Dolby Digital offers no backward compatibility and requires a Dolby decoder in the receiver.
With MPEG Surround, "all broadcasters need to do is to replace a stereo encoder with a surround encoder," said Jan Nordmann, Fraunhofer's marketing manager for audio and multimedia real-time systems. "They can use the existing infrastructure for everything else if they want to introduce multichannel audio to their current audio broadcast."
The technology also lets broadcasters or application providers transmit a compact set of parameters representing the spatial image of the original surround-sound signal along with an automatic mono or stereo downmix. A separate, hand-optimized downmix signal may also be used. While such spatial side information is sent in an auxiliary data field, a receiver's MPEG Surround decoder expands the transmitted downmix signal into a high-quality multichannel output. A receiver with no MPEG Surround decoder ignores the spatial parameter information and decodes the bit stream as stereo. MPEG Surround's surround-enhancement data rate is scalable from 3 to 32 kbits/second.
The MPEG Surround demonstration here--in combination with MPEG-4 HD-AAC--featured good surround-sound audio quality at rates as low as 48 kbits/s. The technology can add multichannel spatial experience "up to discrete quality," according to Fraunhofer's Nordmann.
What's missing is chips. Because the final spec for MPEG Surround is not published, details of decoder processing-power requirements are not available. Grill said that an MPEG Surround audio decoder IC may be "a little more complex" than today's Dolby Digital decoder chips, but that "a high-end DSP available on the market today should be able to handle the task."
The new technology promises a smooth transition from stereo to multichannel sound, since it does not require broadcasters or content owners to send (or store) both stereo and multichannel versions of audio. Potential applications include digital radio systems and Internet streaming. MPEG Surround may also be used to establish multichannel services for mobile-TV broadcasts, according to proponents, and could prove useful in storage media.
See related image
