From Wikipedia, the free encyclopedia - View original article
In computer science, speech recognition (SR) is the translation of spoken words into text. It is also known as "automatic speech recognition", "ASR", "computer speech recognition", "speech to text", or just "STT".
Some SR systems use "speaker independent speech recognition" while others use "training" where an individual speaker reads sections of text into the SR system. These systems analyze the person's specific voice and use it to fine tune the recognition of that person's speech, resulting in more accurate transcription. Systems that do not use training are called "speaker independent" systems. Systems that use training are called "speaker dependent" systems.
Speech recognition applications include voice user interfaces such as voice dialling (e.g. "Call home"), call routing (e.g. "I would like to make a collect call"), domotic appliance control, search (e.g. find a podcast where particular words were spoken), simple data entry (e.g., entering a credit card number), preparation of structured documents (e.g. a radiology report), speech-to-text processing (e.g., word processors or emails), and aircraft (usually termed Direct Voice Input).
The term voice recognition refers to finding the identity of "who" is speaking, rather than what they are saying. Recognizing the speaker can simplify the task of translating speech in systems that have been trained on specific person's voices or it can be used to authenticate or verify the identity of a speaker as part of a security process.
In the health care domain, speech recognition can be implemented in front-end or back-end of the medical documentation process. Front-End speech recognition is where the provider dictates into a speech-recognition engine, the recognized words are displayed as they are spoken, and the dictator is responsible for editing and signing off on the document. Back-End or deferred speech recognition is where the provider dictates into a digital dictation system, the voice is routed through a speech-recognition machine and the recognized draft document is routed along with the original voice file to the editor, where the draft is edited and report finalised. Deferred speech recognition is widely used in the industry currently.
One of the major issues relating to the use of speech recognition in healthcare is that the American Recovery and Reinvestment Act of 2009 (ARRA) provides for substantial financial benefits to physicians who utilize an EMR according to "Meaningful Use" standards. These standards require that a substantial amount of data be maintained by the EMR (now more commonly referred to as an Electronic Health Record or EHR). Unfortunately, in many instances, the use of speech recognition within an EHR will not lead to data maintained within a database, but rather to narrative text. For this reason, substantial resources are being expended to allow for the use of front-end SR while capturing data within the EHR.
Substantial efforts have been devoted in the last decade to the test and evaluation of speech recognition in fighter aircraft. Of particular note is the U.S. program in speech recognition for the Advanced Fighter Technology Integration (AFTI)/F-16 aircraft (F-16 VISTA), and a program in France installing speech recognition systems on Mirage aircraft, and also programs in the UK dealing with a variety of aircraft platforms. In these programs, speech recognizers have been operated successfully in fighter aircraft, with applications including: setting radio frequencies, commanding an autopilot system, setting steer-point coordinates and weapons release parameters, and controlling flight display.
Working with Swedish pilots flying in the JAS-39 Gripen cockpit, Englund (2004) found recognition deteriorated with increasing G-loads. It was also concluded that adaptation greatly improved the results in all cases and introducing models for breathing was shown to improve recognition scores significantly. Contrary to what might be expected, no effects of the broken English of the speakers were found. It was evident that spontaneous speech caused problems for the recognizer, as could be expected. A restricted vocabulary, and above all, a proper syntax, could thus be expected to improve recognition accuracy substantially.
The Eurofighter Typhoon currently in service with the UK RAF employs a speaker-dependent system, i.e. it requires each pilot to create a template. The system is not used for any safety critical or weapon critical tasks, such as weapon release or lowering of the undercarriage, but is used for a wide range of other cockpit functions. Voice commands are confirmed by visual and/or aural feedback. The system is seen as a major design feature in the reduction of pilot workload, and even allows the pilot to assign targets to himself with two simple voice commands or to any of his wingmen with only five commands.
Speaker independent systems are also being developed and are in testing for the F35 Lightning II (JSF) and the Alenia Aermacchi M-346 Master lead-in fighter trainer. These systems have produced word accuracies in excess of 98%.
The problems of achieving high recognition accuracy under stress and noise pertain strongly to the helicopter environment as well as to the jet fighter environment. The acoustic noise problem is actually more severe in the helicopter environment, not only because of the high noise levels but also because the helicopter pilot, in general, does not wear a facemask, which would reduce acoustic noise in the microphone. Substantial test and evaluation programs have been carried out in the past decade in speech recognition systems applications in helicopters, notably by the U.S. Army Avionics Research and Development Activity (AVRADA) and by the Royal Aerospace Establishment (RAE) in the UK. Work in France has included speech recognition in the Puma helicopter. There has also been much useful work in Canada. Results have been encouraging, and voice applications have included: control of communication radios, setting of navigation systems, and control of an automated target handover system.
As in fighter applications, the overriding issue for voice in helicopters is the impact on pilot effectiveness. Encouraging results are reported for the AVRADA tests, although these represent only a feasibility demonstration in a test environment. Much remains to be done both in speech recognition and in overall speech technology in order to consistently achieve performance improvements in operational settings.
Training for air traffic controllers (ATC) represents an excellent application for speech recognition systems. Many ATC training systems currently require a person to act as a "pseudo-pilot", engaging in a voice dialog with the trainee controller, which simulates the dialog that the controller would have to conduct with pilots in a real ATC situation. Speech recognition and synthesis techniques offer the potential to eliminate the need for a person to act as pseudo-pilot, thus reducing training and support personnel. In theory, Air controller tasks are also characterized by highly structured speech as the primary output of the controller, hence reducing the difficulty of the speech recognition task should be possible. In practice, this is rarely the case. The FAA document 7110.65 details the phrases that should be used by air traffic controllers. While this document gives less than 150 examples of such phrases, the number of phrases supported by one of the simulation vendors speech recognition systems is in excess of 500,000.
The USAF, USMC, US Army, US Navy, and FAA as well as a number of international ATC training organizations such as the Royal Australian Air Force and Civil Aviation Authorities in Italy, Brazil, and Canada are currently using ATC simulators with speech recognition from a number of different vendors.
ASR in the field of telephony is now commonplace and in the field of computer gaming and simulation is becoming more widespread. Despite the high level of integration with word processing in general personal computing. However, ASR in the field of document production has not seen the expected[by whom?] increases in use.
The improvement of mobile processor speeds made feasible the speech-enabled Symbian and Windows Mobile Smartphones. Speech is used mostly as a part of User Interface, for creating pre-defined or custom speech commands. Leading software vendors in this field are: Microsoft Corporation (Microsoft Voice Command), Digital Syphon (Sonic Extractor), LumenVox, Nuance Communications (Nuance Voice Control), Speech Technology Center, Vito Technology (VITO Voice2Go), Speereo Software (Speereo Voice Translator), Verbyx VRX and SVOX.
The performance of speech recognition systems is usually evaluated in terms of accuracy and speed. Accuracy is usually rated with word error rate (WER), whereas speed is measured with the real time factor. Other measures of accuracy include Single Word Error Rate (SWER) and Command Success Rate (CSR).
However, speech recognition (by a machine) is a very complex problem. Vocalizations vary in terms of accent, pronunciation, articulation, roughness, nasality, pitch, volume, and speed. Speech is distorted by a background noise and echoes, electrical characteristics. Accuracy of speech recognition vary with the following:
Robot Interaction Language (ROILA) is a constructed language created to address the problems associated with speech interaction using natural languages. ROILA is constructed on the basis of two important goals, firstly it should be learnable by the human user and secondly, the language should be optimized for efficient recognition by a robot.
|This article may need to be rewritten entirely to comply with Wikipedia's quality standards, as section. (June 2012)|
As mentioned earlier in this article, accuracy of speech recognition varies in the following:
e.g. The 10 digits "zero" to "nine" can be recognized essentially perfectly, but vocabulary sizes of 200, 5000 or 100000 may have error rates of 3%, 7% or 45% respectively.
e.g. The 26 letters of the English alphabet are difficult to discriminate because they are confusable words (most notoriously, the E-set: "B, C, D, E, G, P, T, V, Z"); an 8% error rate is considered good for this vocabulary.
A speaker-dependent system is intended for use by a single speaker.
A speaker-independent system is intended for use by any speaker, more difficult.
With isolated speech single words are used, therefore it becomes easier to recognize the speech.
With discontinuous speech full sentences separated by silence are used, therefore it becomes easier to recognize the speech as well as with isolated speech.
With continuous speech naturally spoken sentences are used, therefore it becomes harder to recognize the speech, different from both isolated and discontinuous speech.
e.g. Querying application may dismiss the hypothesis "The apple is red."
e.g. Constraints may be semantic; rejecting "The apple is angry."
e.g. Syntactic; rejecting "Red is apple the."
Constraints are often represented by a grammar.
When a person reads it's usually in a context that has been previously prepared, but when a person uses spontaneous speech, it is difficult to recognize the speech because of the disfluences (like "uh" and "um", false starts, incomplete sentences, stuttering, coughing, and laughter) and limited vocabulary.
Environmental noise (e.g. Noise in a car or a factory)
Acoustical distortions (e.g. echoes, room acoustics)
Speech recognition is a multi-levelled pattern recognition task.
e.g. Phonemes, Words, Phrases, and Sentences;
e.g. Known word pronunciations or legal word sequences, which can compensate for errors or uncertainties at lower level;
By combining decisions probabilistically at all lower levels, and making more deterministic decisions only at the highest level;
Speech recognition by a machine is a process broken into several phases. Computationally, it is a problem in which a sound pattern has to be recognized or classified into a category that represents a meaning to a human. Every acoustic signal can be broken in smaller more basic sub-signals. As the more complex sound signal is broken into the smaller sub-sounds, different levels are created, where at the top level we have complex sounds, which are made of simpler sounds on lower level, and going to lower levels even more, we create more basic and shorter and simpler sounds. The lowest level, where the sounds are the most fundamental, a machine would check for simple and more probabilistic rules of what sound should represent. Once these sounds are put together into more complex sound on upper level, a new set of more deterministic rules should predict what new complex sound should represent. The most upper level of a deterministic rule should figure out the meaning of complex expressions. In order to expand our knowledge about speech recognition we need to take into a consideration neural networks. There are four steps of neural network approaches:
For telephone speech the sampling rate is 8000 samples per second;
computed every 10 ms, with one 10 ms section called a frame;
Analysis of four-step neural network approaches can be explained by further information. Sound is produced by air (or some other medium) vibration, which we register by ears, but machines by receivers. Basic sound creates a wave which has 2 descriptions; Amplitude (how strong is it), and frequency (how often it vibrates per second).
The sound waves can be digitized: Sample a strength at short intervals like in picture above[where?] to get bunch of numbers that approximate at each time step the strength of a wave. Collection of these numbers represent analog wave. This new wave is digital. Sound waves are complicated because they superimpose one on top of each other. Like the waves would. This way they create odd-looking waves. For example, if there are two waves that interact with each other we can add them which creates new odd-looking wave.
Given basic sound blocks, that a machine digitized, one has a bunch of numbers which describe a wave and waves describe words. Each frame has a unit block of sound, which are broken into basic sound waves and represented by numbers after Fourier Transform, can be statistically evaluated to set to which class of sounds it belongs to. The nodes in the figure on a slide represent a feature of a sound in which a feature of a wave from first layer of nodes to a second layer of nodes based on some statistical analysis. This analysis depends on programmer's instructions. At this point, a second layer of nodes represents higher level features of a sound input which is again statistically evaluated to see what class they belong to. Last level of nodes should be output nodes that tell us with high probability what original sound really was.
In 1982, Kurzweil Applied Intelligence and Dragon Systems released speech recognition products. By 1985, Kurzweil’s software had a vocabulary of 1,000 words—if uttered one word at a time. Two years later, in 1987, its lexicon reached 20,000 words, entering the realm of human vocabularies, which range from 10,000 to 150,000 words. But recognition accuracy was only 10% in 1993. Two years later, the error rate crossed below 50%. Dragon Systems released "Naturally Speaking" in 1997, which recognized normal human speech. Progress mainly came from improved computer performance and larger source text databases. The Brown Corpus was the first major database available, containing several million words. In 2006, Google published a trillion-word speech corpus, while Carnegie Mellon University researchers found no significant increase in recognition accuracy.
Both acoustic modeling and language modeling are important parts of modern statistically-based speech recognition algorithms. Hidden Markov models (HMMs) are widely used in many systems. Language modeling has many other applications such as smart keyboard and document classification.
Modern general-purpose speech recognition systems are based on Hidden Markov Models. These are statistical models that output a sequence of symbols or quantities. HMMs are used in speech recognition because a speech signal can be viewed as a piecewise stationary signal or a short-time stationary signal. In a short time-scales (e.g., 10 milliseconds), speech can be approximated as a stationary process. Speech can be thought of as a Markov model for many stochastic purposes.
Another reason why HMMs are popular is because they can be trained automatically and are simple and computationally feasible to use. In speech recognition, the hidden Markov model would output a sequence of n-dimensional real-valued vectors (with n being a small integer, such as 10), outputting one of these every 10 milliseconds. The vectors would consist of cepstral coefficients, which are obtained by taking a Fourier transform of a short time window of speech and decorrelating the spectrum using a cosine transform, then taking the first (most significant) coefficients. The hidden Markov model will tend to have in each state a statistical distribution that is a mixture of diagonal covariance Gaussians, which will give a likelihood for each observed vector. Each word, or (for more general speech recognition systems), each phoneme, will have a different output distribution; a hidden Markov model for a sequence of words or phonemes is made by concatenating the individual trained hidden Markov models for the separate words and phonemes.
Described above are the core elements of the most common, HMM-based approach to speech recognition. Modern speech recognition systems use various combinations of a number of standard techniques in order to improve results over the basic approach described above. A typical large-vocabulary system would need context dependency for the phonemes (so phonemes with different left and right context have different realizations as HMM states); it would use cepstral normalization to normalize for different speaker and recording conditions; for further speaker normalization it might use vocal tract length normalization (VTLN) for male-female normalization and maximum likelihood linear regression (MLLR) for more general speaker adaptation. The features would have so-called delta and delta-delta coefficients to capture speech dynamics and in addition might use heteroscedastic linear discriminant analysis (HLDA); or might skip the delta and delta-delta coefficients and use splicing and an LDA-based projection followed perhaps by heteroscedastic linear discriminant analysis or a global semitied covariance transform (also known as maximum likelihood linear transform, or MLLT). Many systems use so-called discriminative training techniques that dispense with a purely statistical approach to HMM parameter estimation and instead optimize some classification-related measure of the training data. Examples are maximum mutual information (MMI), minimum classification error (MCE) and minimum phone error (MPE).
Decoding of the speech (the term for what happens when the system is presented with a new utterance and must compute the most likely source sentence) would probably use the Viterbi algorithm to find the best path, and here there is a choice between dynamically creating a combination hidden Markov model, which includes both the acoustic and language model information, and combining it statically beforehand (the finite state transducer, or FST, approach).
A possible improvement to decoding is to keep a set of good candidates instead of just keeping the best candidate, and to use a better scoring function (rescoring) to rate these good candidates so that we may pick the best one according to this refined score. The set of candidates can be kept either as a list (the N-best list approach) or as a subset of the models (a lattice). Rescoring is usually done by trying to minimize the Bayes risk (or an approximation thereof): Instead of taking the source sentence with maximal probability, we try to take the sentence that minimizes the expectancy of a given loss function with regards to all possible transcriptions (i.e., we take the sentence that minimizes the average distance to other possible sentences weighted by their estimated probability). The loss function is usually the Levenshtein distance, though it can be different distances for specific tasks; the set of possible transcriptions is, of course, pruned to maintain tractability. Efficient algorithms have been devised to rescore lattices represented as weighted finite state transducers with edit distances represented themselves as a finite state transducer verifying certain assumptions.
Dynamic time warping is an approach that was historically used for speech recognition but has now largely been displaced by the more successful HMM-based approach.
Dynamic time warping is an algorithm for measuring similarity between two sequences that may vary in time or speed. For instance, similarities in walking patterns would be detected, even if in one video the person was walking slowly and if in another he or she were walking more quickly, or even if there were accelerations and decelerations during the course of one observation. DTW has been applied to video, audio, and graphics – indeed, any data that can be turned into a linear representation can be analyzed with DTW.
A well-known application has been automatic speech recognition, to cope with different speaking speeds. In general, it is a method that allows a computer to find an optimal match between two given sequences (e.g., time series) with certain restrictions. That is, the sequences are "warped" non-linearly to match each other. This sequence alignment method is often used in the context of hidden Markov models.
Neural networks emerged as an attractive acoustic modeling approach in ASR in the late 1980s. Since then, neural networks have been used in many aspects of speech recognition such as phoneme classification, isolated word recognition, and speaker adaptation.
In contrast to HMMs, neural networks make no assumptions about feature statistical properties and have several qualities making them attractive recognition models for speech recognition. When used to estimate the probabilities of a speech feature segment, neural networks allow discriminative training in a natural and efficient manner. Few assumptions on the statistics of input features are made with neural networks. However, in spite of their effectiveness in classifying short-time units such as individual phones and isolated words, neural networks are rarely successful for continuous recognition tasks, largely because of their lack of ability to model temporal dependencies. Thus, one alternative approach is to use neural networks as a pre-processing e.g. feature transformation, dimensionality reduction, for the HMM based recognition.
Popular speech recognition conferences held each year or two include SpeechTEK and SpeechTEK Europe, ICASSP, Interspeech/Eurospeech, and the IEEE ASRU. Conferences in the field of natural language processing, such as ACL, NAACL, EMNLP, and HLT, are beginning to include papers on speech processing. Important journals include the IEEE Transactions on Speech and Audio Processing (now named IEEE Transactions on Audio, Speech and Language Processing), Computer Speech and Language, and Speech Communication.
Books like "Fundamentals of Speech Recognition" by Lawrence Rabiner can be useful to acquire basic knowledge but may not be fully up to date (1993). Another good source can be "Statistical Methods for Speech Recognition" by Frederick Jelinek and "Spoken Language Processing (2001)" by Xuedong Huang etc. More up to date is "Computer Speech", by Manfred R. Schroeder, second edition published in 2004. The recently updated textbook of "Speech and Language Processing (2008)" by Jurafsky and Martin presents the basics and the state of the art for ASR. Speaker recognition also uses the same features, most of the same front-end processing, and classification techniuqes as is done in speech recognition. A most recent comprehensive textbook, "Fundamentals of Speaker Recognition" by Homayoon Beigi, is an in depth source for up to date details on the theory and practice. A good insight into the techniques used in the best modern systems can be gained by paying attention to government sponsored evaluations such as those organised by DARPA (the largest speech recognition-related project ongoing as of 2007 is the GALE project, which involves both speech recognition and translation components).
A good and accessible introduction to speech recognition technology and its history is provided by the general audience book "The Voice in the Machine. Building Computers That Understand Speech" by Roberto Pieraccini (2012)
In terms of freely available resources, Carnegie Mellon University's Sphinx toolkit is one place to start to both learn about speech recognition and to start experimenting. Another resource (free as in free beer, not as in free speech) is the HTK book (and the accompanying HTK toolkit). The AT&T libraries GRM and DCD are also general software libraries for large-vocabulary speech recognition.
For more software resources, see List of speech recognition software.
A useful review of the area of robustness in ASR is provided by Junqua and Haton (1995).
People with disabilities can benefit from speech recognition programs. For individuals that are Deaf or Hard of Hearing, speech recognition software is used to automatically generate a closed-captioning of conversations such as discussions in conference rooms, classroom lectures, and/or religious services.
Speech recognition is also very useful for people who have difficulty using their hands, ranging from mild repetitive stress injuries to involved disabilities that preclude using conventional computer input devices. In fact, people who used the keyboard a lot and developed RSI became an urgent early market for speech recognition. Speech recognition is used in deaf telephony, such as voicemail to text, relay services, and captioned telephone. Individuals with learning disabilities who have problems with thought-to-paper communication (essentially they think of an idea but it is processed incorrectly causing it to end up differently on paper) can possibly benefit from the software but the technology is not bug proof. Also the whole idea of speak to text can be hard for intellectually disabled person's due to the fact that it is rare that anyone tries to learn the technology to teach the person with the disability.
This type of technology can help those with dyslexia but other disabilities are still in question. The effectiveness of the product is the problem that is hindering it being effective. Although a kid may be able to say a word depending on how clear they say it the technology may think they are saying another word and input the wrong one. Giving them more work to fix, causing them to have to take more time with fixing the wrong word.
Measuring progress in speech recognition performance is difficult and controversial. Some speech recognition tasks are much more difficult than others. Word error rates on some tasks are less than 1%. On others they can be as high as 50%. Sometimes it even appears that performance is going backward, as researchers undertake harder tasks that have higher error rates.
Because progress is slow and is difficult to measure, there is some perception that performance has plateaued and that funding has dried up or shifted priorities. Such perceptions are not new. In 1969, John Pierce wrote an open letter that did cause much funding to dry up for several years. In 1993 there was a strong feeling that performance had plateaued and there were workshops dedicated to the issue. However, in the 1990s, funding continued more or less uninterrupted and performance continued, slowly but steadily, to improve.
For the past thirty years, speech recognition research has been characterized by the steady accumulation of small incremental improvements. There has also been a trend to change focus to more difficult tasks due both to progress in speech recognition performance and to the availability of faster computers. In particular, this shifting to more difficult tasks has characterized DARPA funding of speech recognition since the 1980s. In the last decade, it has continued with the EARS project, which undertook recognition of Mandarin and Arabic in addition to English, and the GALE project, which focused solely on Mandarin and Arabic and required translation simultaneously with speech recognition.
Commercial research and other academic research also continue to focus on increasingly difficult problems. One key area is to improve robustness of speech recognition performance, not just robustness against noise but robustness against any condition that causes a major degradation in performance. Another key area of research is focused on an opportunity rather than a problem. This research attempts to take advantage of the fact that in many applications there is a large quantity of speech data available, up to millions of hours. It is too expensive to have humans transcribe such large quantities of speech, so the research focus is on developing new methods of machine learning that can effectively utilize large quantities of unlabeled data. Another area of research is better understanding of human capabilities and to use this understanding to improve machine recognition performance.
|editor3-given=ignored (help); Unknown parameter
|editor4-given=ignored (help); More than one of