What is Speech Recognition?

Computers equipped with a source of sound input can interpret human speech through speech recognition technologies.

Since the 1990s, speech recognition commercial systems have been available off-the-shelf. Even though this technology was a success, very few people use such speech recognition systems on their desktop computers. Despite the fact that most people are able to speak considerably faster than they can type, they find conventional input devices to be more convenient to interact with their computers. However, in some cases, using both keyboard and speech recognition simultaneously can be more efficient than using any one of these inputs alone. Current speech recognition technologies are inefficient in an office environment with high amplitude of background speech. Large-vocabulary systems with speaker-independence that are designed to operate within these adverse environments have significantly lower recognition accuracy. The typical achievable recognition rate as of 2005 for large-vocabulary speaker-independent systems is about 80%-90% for a clear environment, but can be as low as 50% for scenarios like mobile phones with background noise. Additionally, heavy use of the speech organs can result in vocal loading.

Speech recognition systems are useful where the speed of text input is required to be extremely fast. They are used in legal and medical transcription, the generation of subtitles for live sports and current affairs programs on television; not directly but via an operator that re-speaks the dialog into software trained in the operator’s voice. In courtrooms and similar situations where the operator’s voice would disturb the proceedings, he or she may sit in a soundproofed booth or wear a Stenomask or similar device.

For people who have difficulty interacting with their computers through a keyboard, such as those with serious carpal tunnel syndrome, injured hands or arms or other physical limitations, may find speech recognition to be a necessity.

Speech recognition technology is used mostly for telephone applications like travel booking and information, financial account information, customer service call routing and directory assistance. Using constrained grammar recognition applications can can achieve remarkably high accuracy. The research and development in speech recognition technology is continually increasing as the cost for implementing such voice-activated systems has dropped and the usefulness and efficacy of these systems has improved. Speech recognition has also enabled the automation of certain applications that are not automatable using push-button interactive voice response (IVR) systems. However, speech recognition based systems remain the exception as push-button systems are still much more economical to implement and operate.

Source: Wikipedia