Digital recordings can be made by speaking into a phone, a microphone plugged directly into a computer or a hand-held unit. The recorded files can then be transferred via modem, a disk-like card, e-mail or network, and stored on computer or network.
With speech to text software, transcription is virtually instantaneous. The task becomes nothing more than proofreading the converted text file against the original voice file. Philips Electronics offers natural continuous speech recognition software customized for each discipline before it's introduced, says Linda Dunlea, Philips marketing and sales support director. Medical and legal versions are on the market, and software for banking and insurance is in the works. Future releases will be geared toward law enforcement and government. Dragon Systems recently released general purpose continuous speech recognition software targeted at small offices.
Other technological advances include clearer, more realistic recordings. A telephone transmits sound in the 300 to 3,000 hertz range, according to Sony marketing manager Steve Teplansky. His company's newest units capture sound at 10 to 15,000 hertz, "almost exceeding the human hearing range," he says.
NEXT STORY: Transformation of Quality Efforts