At Mobile World Congress (MWC) this week, handset vendors such as HTC, LG, and ZTE have rolled out their latest generations of phones supporting HD voice on HSPA+ and soon LTE (News - Alert). Underneath the hood of the hottest hardware are a combination of silicon and software stacks to deliver optimum voice quality in HD voice and other environments.
Perhaps the key buzz-phrase out of MWC this year is SRVCC -- Single Radio Voice Call Continuity. SRVCC allows HD voices to move back and forth between LTE networks and 3G GSM-type networks with (yes, you guessed it) a single radio part, rather than having one radio talking to an LTE network and a second one talking to the 3G network. Single radio means fewer parts, less complexity, lower cost, lower power consumption, longer battery life and happy customers who can keep chatting in HD voice as they move between networks rather than getting bumped down to narrowband.
NVIDIA has received a lot of attention for its quad-core Tegra 3 processor (It's actually four cores plus a low-power core, but that's a story for another day). However, it has offered the Tegra 3 with its Icera (News - Alert) 410 LTE modem chip and is talking up the virtues of SRVCC in phones using the two chips.
However, silicon is only as good as the software that supports it. This week D2 Technologies (News - Alert) revealed its role in the NVIDIA/Icera SRVCC story by providing the 4G version of its mCUE VoIP stack to tie everything together. Handset OEMs get all the essentials -- CPU, radio, software stack -- to implement SRVCC on Android phones.
A number of high-end phones incorporate Audience's (News - Alert) voice and audio processing chip to provide clearer voice quality during phone calls, even in noisy environments. Being able to provide echo cancellation and suppress background noise is essential for HD voice calls to sound like voice calls rather than passing along everything captured by the microphone, including the people talking behind you, speeding cars and the like.
Audience has now introduced a voice processing chip for low-cost feature phones, the eS110. The audio DSP core provides stationary and non-stationary noise suppression, acoustic echo cancellation, and a "HiFi Voice" feature to provide an voice upconvert experience for narrowband calls to make them sound more like wideband HD voice; the last feature is likely to add a bit more confusion in the growing marketing speak between wideband, superwideband, Full-HD voice from Fraunhofer (News - Alert), and the misnomer of HD audio.
Most of this technology is already available and being incorporated in phones to roll out later this year for LTE deployments around the world. For example, a number of phones being showcased with NVIDA's Tegra 3 processor and an HSPA+ modem chip will show up later this year with LTE chips added.
Edited by Rich Steeves