Wideband Audio Technology Featured Article
July 25, 2008
HD Voice: Raising the Bar on Phone Call Quality
Perhaps the biggest frustration associated with modern phone communications is mediocre voice quality. Just about everyone at some point has experienced the annoyance and time-wastage of a poor voice connection, during which one or more of the parties conversing cannot understand one another, or have a difficult time doing so.
Just as technology introduced the problem of voice quality, it also can solve it. Technologies now exist that can be categorized as “high definition voice,” or HD voice. The need for development and deployment of HD voice telephony is being driven by the bar having been raised on other multimedia applications — like HD television, HD radio, and HD discs (think HD DVD, Blu-ray).
While the quality of telephony voice hasn’t changed much (except perhaps to get a bit worse with the introduction of cell phones) in a while (PSTN speech encoding has been the same for 50 years), consumers are now demanding an upgrade and the industry is responding.
Many communications vendors are now improving the quality of voice carried over IP networks (think VoIP) by using industry standard, wideband codecs for more lifelike audio quality during phone calls.
In a recent white paper, “Wideband Audio and IP Telephony: Experience Higher-Quality Media,” Cisco (News - Alert) pointed out that, while the standard with improving newer telephony technologies (e.g. VoIP) is to match the “toll quality voice” standard (the quality of traditional, PSTN phone calls), HD voice actually improves upon the older standard to truly raise the bar on voice communications.
Toll quality voice, Cisco said in its white paper, “is mediocre at best when compared to other sources of audio transmission.” What constitutes good and excellent voice quality can be defined numerically. The human voice produces sound frequencies ranging from 30 to 18,0000 Hz. Most of intelligible speech is contained in the 0-400 Hz band, which is defined as the “voice channel.” Eliminating noise from voice for phone communications limits the range to 300-3400 Hz.
Even more specifically, Cisco said in its white paper, “analysis of the frequency spectrum of human speech shows that much of the differentiating content in certain sounds is found in the 4- to 18-kHz range.” Translating these numbers to digital voice, which has been in use since the 1960s, is a somewhat complex process. Converting a continuous analog signal into a discrete digital signal led to the development of pulse code modulation (PCM), “which takes the signal level at the time of sampling and converts it into a
digital value to represent the signal.”
The result of all this analysis was the establishment of the 8000 times/second frequency sampling rate, which equals 64,000 bps or 64 kbps. Today, the 64-kbps PCM encoded voice stream is considered standard for “toll quality.”
However, modern HD voice technology can do much better than that.
Wideband voice quality, or HD voice, is usually defined by referring to the G.722 codec, which supports a much wider voice frequency range but uses the same 64-kbps stream as toll quality PCM. How much wider? G.722 offers a frequency range of 8 kHz.
G.722, Cisco explained in its white paper, puts voice signal through a digital filter to separate it into two bands: 0-4 kHz and 4-8 kHz. These bands are then encoded using Adaptive Differential PCM (ADPCM), with most of the voice energy taking up 48 kbps of bandwidth and the remaining 16 kbps used for the higher sub-band.
“By performing ADPCM encoding on each sub-band separately, G.722 can provide both low and high frequencies that will provide richer audio sound and better re-creation of the original signal,” Cisco explained in its white paper.
All of this might seem complex, because under the hood it is. But for the average user, the take-away knowledge should simply be that, as good as traditional telephone audio was, modern HD voice technology is dramatically improving upon that quality. In other words, HD voice means phone calls can now, and in the future will be, clearer than they’ve ever been.
Mae Kowalke is senior editor for TMCnet, covering VoIP, CRM, call center and wireless technologies. To read more of Mae’s articles, please visit her columnist page. She also blogs for TMCnet here.
Don’t forget to check out TMCnet’s White Paper Library, which provides a selection of in-depth information on relevant topics affecting the IP Communications industry. The library offers white papers, case studies and other documents which are free to registered users. Today’s featured white paper is Jim Cossetta, President, CEO, 4What Interactive, Creators of The VoIPTrainer, brought to you by 4What Interactive (News - Alert).
TMCnet LOGIN
SUBSCRIPTIONS
By 

