Voice over LTE (VoLTE) is barely out of the gate, but Ericsson (News - Alert) is pushing the Next Big Thing. "Evolved HD voice for LTE" will be a "new mobile voice experience," says the company in a recent blog post and white paper release.
The "evolved" HD voice solution comes out of a new 3GPP codec standard called Enhanced Voice Services (EVS), incorporating support for superwideband (SWB) and Fullband (FB). To get it, operators have to have the underlying infrastructure for VoLTE up and running. EVS drops in as the codec, replacing AMR-WB as well as providing backwards compatibility with AMR-WB and AMR. What this means is EVS can still do basic HD voice calling with existing 2G and 3G networks using AMR-WB for HD voice as well with VoLTE networks that haven't implemented EVS.
EVS supports the full frequency range of human voice, from around 50 Hertz (Hz) to over 14000 Hz. Narrowband PSTN calls and cellular deliver voice in a range of 300 Hz to 3400 Hz while AMR-WB and G.722 go up to 7000 Hz. The wider frequency range provides EVS the ability to deliver much improved music quality if the bandwidth is available.
Speaking of bandwidth, EVS also delivers better audio performance while using less bandwidth, with a BVR mode capable of delivering narrowband and stock HD voice in about 5.95 kbps to 7.36 kbps and superwideband starting at 13.2 kbps, depending on what modes the carrier configures for handsets and network. AMR-WB in HD voice mode starts at around 12 kbps, and can be configured to use up to 23.85 kbps, as T-Mobile (News - Alert) US does for VoLTE.
Less bandwidth while delivering better voice quality performance is always a choice that carriers want, since they always want to get more data out of existing spectrum licensing. By offering "superwideband" mode at 13.2 kbps, carriers could provide a "better than (AMR-WB) HD voice" experience with a little more bandwidth.
Among the 13 contributors to this codec are the usual suspects, including Ericsson, Fraunhofer, Qualcomm (News - Alert). There's no discussion on how/if licensing of EVS to third-parties will be handled, but it is bound to be a royal pain if the process is similar to AMR-WB, with the requirement to go to each separate patent contributor to work out terms. Ericsson's white paper says commercial deployments "could be expected" in 2015, but the licensing issues may delay third-party support. Judging from the discussions I had last month at CTIA (News - Alert), Ericsson and others have little sympathy for third-party IP (intellectual property) licensing headaches
Another potential for delay is the rollout of VoLTE networks at large. Worldwide, VoLTE will only be commercially available on 19 networks by the end of this year, reaching 38 by 2015, according to estimates by the Global Mobile Suppliers Association (GSA). Verizon (News - Alert) has been struggling with its VoLTE rollout, so I don't see them rushing into deploying a new codec anytime soon.
On the other hand, I'm willing to bet T-Mobile US is in talks with Ericsson to roll out the new codec, given the relationship between the two companies and T-Mobile's love to challenge the status quo. T-Mobile already has the best HD voice service coverage in the U.S., so it's a short leap for them to drop EVS into their VoLTE infrastructure. Ericsson and the rest of the EVS bunch could offer some incentives (discounts) to T-Mobile to use the new codec operationally.
Ericsson offers three major justifications to run HD voice and evolved HD voice: More and longer calls, attractive smartphone bundles, and enterprise benefit from high-quality voice services. The "more and longer calls" theme only works if operators are using a price-per-minute model, but the logic backfires a bit due to the proliferation of over-the-top (OTT) services and a shift from voice to multi-mode communications, including text , email, and social media. "Attractive smartphone bundles" is a euphemism for product differentiation , and it only works over the short term because ultimately all carriers end up offering HD voice service within a market.
The only reasoning which works to promote higher quality voice services: Enterprises—and businesses—can benefit. Voice is information and there's range of existing services from Ye Olde Voice Conferencing to voice biometrics, hypervoice, and voice analytics that benefit from higher quality audio information delivered to people and machine processes. Carriers need to recognize their own services and business processes will benefit significantly from higher quality voice, but such logic escapes them.