speech recognition


Also found in: Dictionary, Medical, Acronyms, Wikipedia.

speech recognition

[′spēch ‚rek·ig′nish·ən]
(engineering acoustics)
The process of analyzing an acoustic speech signal to identify the linguistic message that was intended, so that a machine can correctly respond to spoken commands.
McGraw-Hill Dictionary of Scientific & Technical Terms, 6E, Copyright © 2003 by The McGraw-Hill Companies, Inc.

speech recognition

(application)
(Or voice recognition) The identification of spoken words by a machine. The spoken words are digitised (turned into sequence of numbers) and matched against coded dictionaries in order to identify the words.

Most systems must be "trained," requiring samples of all the actual words that will be spoken by the user of the system. The sample words are digitised, stored in the computer and used to match against future words. More sophisticated systems require voice samples, but not of every word. The system uses the voice samples in conjunction with dictionaries of larger vocabularies to match the incoming words. Yet other systems aim to be "speaker-independent", i.e. they will recognise words in their vocabulary from any speaker without training.

Another variation is the degree with which systems can cope with connected speech. People tend to run words together, e.g. "next week" becomes "neksweek" (the "t" is dropped). For a voice recognition system to identify words in connected speech it must take into account the way words are modified by the preceding and following words.

It has been said (in 1994) that computers will need to be something like 1000 times faster before large vocabulary (a few thousand words), speaker-independent, connected speech voice recognition will be feasible.
This article is provided by FOLDOC - Free Online Dictionary of Computing (foldoc.org)

voice recognition

(1) Using a person's voice as a form of identification. See two-factor authentication.

(2) The conversion of spoken words into computer text. Speech is first digitized and then matched against a dictionary of coded waveforms. Also called "speech recognition," the matches are converted into text as if the words were typed on the keyboard. "Speaker-dependent" systems require users to enunciate samples to train and fine tune the system. "Speaker-independent" recognition such as telephone voice response systems do not require training but generally handle only a limited vocabulary.

Three Categories
The least taxing on the electronics, "command" systems recognize several dozen words and eliminate using the mouse or keyboard. "Discrete voice" recognition systems used for dictation require a pause between each word. "Continuous voice" recognition understands natural speech without pauses and is the most process intensive. The Holy Grail of voice recognition, speaker-independent, continuous systems that handle extensive vocabularies are slowly but surely becoming mainstream. Contrast with speaker recognition.


First Handheld Speech Recognition
The first continuous dictation in a handheld device was in 2000 when Lernout & Hauspie showed off this Linux PDA prototype. It provided keyboard-free email composition. (Image courtesy of Lernout & Hauspie.)
Copyright © 1981-2019 by The Computer Language Company Inc. All Rights reserved. THIS DEFINITION IS FOR PERSONAL USE ONLY. All other reproduction is strictly prohibited without permission from the publisher.
References in periodicals archive ?
AI-based voice and speech recognition software is projected to witness a high CAGR during the forecast period owing to continual development of machine learning techniques and integration of connected devices with personal assistants.
Apple's voice-recognition system, which integrates with Siri, offers on-device speech recognition for some languages, including English, but the system also adjusts to the population as a whole over time.
Voicebrook was founded in 2002 by Pathology and Speech Recognition industry veterans and is based in Long Island, New York.
With the rise of Amazon's Alexa, Google's Assistant, Apple's Siri (which is based on Nuance's speech recognition), and Microsoft's Cortana, contextual understanding leapt forward due to the billions of utterances constrained in boundaries such as maps and directions, computer commands (e.g., "open Word" or "send text message"), automotive commands, etc.
"We are proud to see Arabic speech recognition is a world-class technology and mature enough to be trusted in the news room by world leading news organisations.
The free trial includes 10 free speech recognition as well as 10 free transcription service minutes, so users can test these two additional services as well.
The challenge of building a speech recognition engine for audio announcements mainly lies in the lack of data.
Speech recognition scores upon delivery of time-compressed sentences under both quiet and noisy conditions and gap detection thresholds were measured and compared between HF SNHL groups with the same cutoff frequency but various degrees of HF SNHL and age-matched NH group.
And every use of the solution contributes to even more accurate speech recognition results, he adds.
Their design is generally optimized for speech recognition and phone call quality and they tend to band limit the signal.
The researchers tested the speech recognition system on the "Switchboard" speech recognition system.

Full browser ?