Speech recognition

Last updated

Speech recognition (automatic speech recognition (ASR), computer speech recognition, or speech-to-text (STT)) is a sub-field of computational linguistics concerned with methods and technologies that translate spoken language into text or other interpretable forms. [1]

Contents

Speech recognition applications include voice user interfaces, where the user speaks to a device, which “listens” and processes the audio. Common voice applications include interpreting commands for calling, call routing, home automation, and aircraft control. This is called direct voice input. Productivity applications including searching audio recordings, creating transcripts, and dictation.

Speech recognition can be used to analyse speaker characteristics, such as identifying native language using pronunciation assessment. [2]

Voice recognition [3] [4] [5] (speaker identification) [6] [7] [8] refers to identifying the speaker, rather than speech contents. Recognizing the speaker can simplify the task of translating speech in systems trained on a specific person's voice. It can also be used to authenticate the speaker as part of a security process.

History

Speech recognition developed over many decades, with progress accelerated due to advances in deep learning and the use of big data. These advances are reflected in an increase in academic papers, [9] and greater system adoption. [10]

Key areas of growth include vocabulary size, more accurate recognition for unfamiliar speakers (speaker independence), and faster processing speed.

Pre-1970

Raj Reddy was the first person to work on continuous speech recognition,[ citation needed ] as a graduate student at Stanford University in the late 1960s. Previous systems required users to pause after each word. Reddy's system issued spoken commands for playing chess.

Around this time, Soviet researchers invented the dynamic time warping (DTW) algorithm[ citation needed ] and used it to create a recognizer capable of operating on a 200-word vocabulary. [17] DTW processed speech by dividing it into short frames (e.g. 10 ms segments) and treating each frame as a unit. Speaker independence, however, remained unsolved.

1970–1990

During the late 1960s, Leonard Baum developed the mathematics of Markov chains at the Institute for Defense Analysis. A decade later, at CMU, Raj Reddy's students James Baker and Janet M. Baker began using the hidden Markov model (HMM) for speech recognition. [22] James Baker had learned about HMMs while at the Institute for Defense Analysis. [23] HMMs enabled researchers to combine sources of knowledge, such as acoustics, language, and syntax, in a unified probabilistic model.

By the mid-1980s, Fred Jelinek's team at IBM created a voice-activated typewriter called Tangora, which could handle a 20,000-word vocabulary. [24] Jelinek's statistical approach placed less emphasis on emulating human brain processes in favor of statistical modelling. (Jelinek's group independently discovered the application of HMMs to speech. [23] ) This was controversial among linguists since HMMs are too simplistic to account for many features of human languages. [25] However, the HMM proved to be a highly useful way for modelling speech and replaced dynamic time warping as the dominant speech recognition algorithm in the 1980s. [26] [27]

Practical speech recognition

The 1980s also saw the introduction of the n-gram language model.

At the end of the DARPA program in 1976, the best computer available to researchers was the PDP-10 with 4 MB of RAM. [30] It could take up to 100 minutes to decode 30 seconds of speech. [31]

Practical products included:

By the early 1990s, the vocabulary of the typical commercial speech recognition system had exceeded the average human vocabulary. [30] Reddy's former student, Xuedong Huang, developed the Sphinx-II system at CMU. Sphinx-II was the first to do speaker-independent, large vocabulary, continuous speech recognition, and it won DARPA's 1992 evaluation. Handling continuous speech with a large vocabulary was a major milestone. Huang later founded the speech recognition group at Microsoft in 1993. Reddy's student Kai-Fu Lee joined Apple, where, in 1992, he helped develop the Casper speech interface prototype.

Lernout & Hauspie, a Belgium-based speech recognition company, acquired other companies, including Kurzweil Applied Intelligence in 1997 and Dragon Systems in 2000. L&H was used in Windows XP. L&H was an industry leader until an accounting scandal destroyed it in 2001. L&H speech technology was bought by ScanSoft, which became Nuance in 2005. Apple licensed Nuance software for its digital assistant Siri. [36]

2000s

In the 2000s, DARPA sponsored two speech recognition programs: Effective Affordable Reusable Speech-to-Text (EARS) in 2002, followed by Global Autonomous Language Exploitation (GALE) in 2005. Four teams participated in EARS: IBM; a team led by BBN with LIMSI and the University of Pittsburgh; Cambridge University; and a team composed of ICSI, SRI, and the University of Washington. EARS funded the collection of the Switchboard telephone speech corpus, which contained 260 hours of recorded conversations from over 500 speakers. [37] The GALE program focused on Arabic and Mandarin broadcast news. Google's first effort at speech recognition came in 2007 after recruiting Nuance researchers. [38] Its first product, GOOG-411, was a telephone-based directory service.

Since at least 2006, the U.S. National Security Agency has employed keyword spotting, allowing analysts to index large volumes of recorded conversations and identify speech containing "interesting" keywords. [39] Other government research programs focused on intelligence applications, such as DARPA's EARS program and IARPA's Babel program.

In the early 2000s, speech recognition was dominated by hidden Markov models combined with feed-forward artificial neural networks (ANN). [40] Later, speech recognition was taken over by long short-term memory (LSTM), a recurrent neural network (RNN) published by Sepp Hochreiter & Jürgen Schmidhuber in 1997. [41] LSTM RNNs avoid the vanishing gradient problem and can learn "Very Deep Learning" tasks [42] that require memories of events that happened thousands of discrete time steps earlier, which is important for speech.

Around 2007, LSTMs trained with Connectionist Temporal Classification (CTC) [43] began to outperform. [44] In 2015, Google reported a 49 percent error‑rate reduction in its speech recognition via CTC‑trained LSTM. [45] Transformers, a type of neural network based solely on attention, were adopted in computer vision [46] [47] and language modelling, [48] [49] and then to speech recognition. [50] [51] [52]

Deep feed-forward (non-recurrent) networks for acoustic modelling were introduced in 2009 by Geoffrey Hinton and his students at the University of Toronto, and by Li Deng [53] and colleagues at Microsoft Research. [54] [55] [56] [57] In contrast to the prioer incremental improvements, deep learning decreased error rates by 30%. [57]

Both shallow and deep forms (e.g., recurrent nets) of ANNs had been explored since the 1980s. [58] [59] [60] However, these methods never defeated non-uniform internal-handcrafting Gaussian mixture model/hidden Markov model (GMM-HMM) technology. [61] Difficulties analyzed in the 1990s, included gradient diminishing [62] and weak temporal correlation structure. [63] [64] All these difficulties combined with insufficient training data and computing power. Most speech recognition pursued generative modelling approaches until deep learning won the day. Hinton et al. and Deng et al. [55] [56] [65] [66]

2010s

By early the 2010s, speech recognition [67] [68] [69] was differentiated from speaker recognition, and speaker independence was considered a major breakthrough. Until then, systems required a "training" period for each voice. [14]

In 2017, Microsoft researchers reached the human parity milestone of transcribing conversational speech on the widely benchmarked Switchboard task. Multiple deep learning models were used to improve accuracy. The error rate was reported to be as low as 4 professional human transcribers working together on the same benchmark. [70]

Models, methods, and algorithms

Both acoustic modelling and language modelling are important parts of statistically-based speech recognition algorithms. Hidden Markov models (HMMs) are widely used in many systems. Language modelling is also used in many other natural language processing applications, such as document classification or statistical machine translation.

Hidden Markov models

Speech recognition systems are based on HMMs. 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 scale (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.

HMMs are popular because they can be trained automatically and are simple and computationally feasible. An HMM outputs a sequence of n-dimensional real-valued vectors (where n is an integer such as 10), outputting one every 10 milliseconds. The vectors consist of cepstral coefficients, obtained by a Fourier transform of a short window of speech and decorrelating the spectrum using a cosine transform, then taking the first (most significant) coefficients. The HMM tends to have, in each state, a statistical distribution that is a mixture of diagonal covariance Gaussians, which give a likelihood for each observed vector. Each word, or (for more general speech recognition systems), each phoneme, has a different output distribution; an HMM for a sequence of words or phonemes is made by concatenating the individual trained HMMs for the separate words and phonemes.

Speech recognition systems use combinations of standard techniques to improve results. A typical large-vocabulary system applies context dependency for the phonemes (so that phonemes with different left and right context have different realizations as HMM states). It uses cepstral normalization to handle speaker and recording conditions. It might use vocal tract length normalization (VTLN) for male-female normalization and maximum likelihood linear regression (MLLR) for more general adaptation. The features use delta and delta-delta coefficients to capture speech dynamics, and in addition might use heteroscedastic linear discriminant analysis (HLDA); or might use splicing and LDA-based projection, followed by HLDA or a global semi-tied covariance transform (also known as maximum likelihood linear transform (MLLT)). Many systems use 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).

Dynamic time warping (DTW)-based speech recognition

Dynamic time warping was historically used for speech recognition, but was later displaced by HMM.

Dynamic time warping measures similarity between two sequences that may vary in time or speed. For instance, similarities in walking patterns could be detected, even if in one video a person was walking slowly and in another was walking more quickly, or even if accelerations and decelerations came during one observation. DTW has been applied to video, audio, and graphics – any data that can be turned into a linear representation can be analyzed with DTW.

This could handle speech at different speaking speeds. In general, it allows an optimal match between two sequences (e.g., time series) with certain restrictions. The sequences are "warped" non-linearly to match each other. This sequence alignment method is often used in the context of HMMs.

Neural networks

Neural networks became interesting in the late 1980s before beginning to dominate in the 2010s. Neural networks have been used in many aspects of speech recognition, such as phoneme classification, [71] phoneme classification through multi-objective evolutionary algorithms, [72] isolated word recognition, [73] audiovisual speech recognition, audiovisual speaker recognition, and speaker adaptation.

Neural networks make fewer explicit assumptions about feature statistical properties than HMMs. When used to estimate the probabilities of a speech segment, neural networks allow natural and efficient discriminative training. However, in spite of their effectiveness in classifying short-time units such as individual phonemes and isolated words, [74] early neural networks were rarely successful for continuous recognition because of their limited ability to model temporal dependencies.

One approach was to use neural networks for feature transformation, or dimensionality reduction. [75] However, more recently, LSTM and related recurrent neural networks (RNNs), [41] [45] [76] [77] Time Delay Neural Networks (TDNN's), [78] and transformers [50] [51] [52] demonstrated improved performance.

Deep feedforward and recurrent neural networks

Researchers are exploring deep neural networks (DNNs) and denoising autoencoders [79] .A DNN is a type of artificial neural network that includes multiple hidden layers between the input and output. [55] Like simpler neural networks, DNNs can model complex, non-linear relationships. However, their deeper architecture allows them to build more sophisticated representations that combine features from earlier layers. This gives them a powerful ability to learn and recognize complex patterns in speech data. [80]

A major breakthrough in using DNNs for large vocabulary speech recognition came in 2010. In a collaboration between industry and academia, researchers used DNNs with large output layers based on context-dependent HMM states that were created using decision trees. [81] [82] [83] This approach significantly improved performanc. [84] [85] [86]

A core idea behind deep learning is to eliminate the need for manually designed features and instead learn directly from input data. This was first demonstrated using deep autoencoders trained on raw spectrograms or linear filter-bank features. [87] These models outperformed traditional Mel-Cepstral features, which rely on fixed transformations. More recently, researchers showed that waveforms can achieve excellent results in large-scale speech recognition. [88]

End-to-end learning

Since 2014, much research has considered "end-to-end" ASR. Traditional phonetic-based (i.e., all HMM-based model) approaches required separate components and training for pronunciation, acoustic, and language. End-to-end models learn from all the components at once. This simplifies the training and deployment processes. For example, an n-gram language model is required for all HMM-based systems, and a typical 2025-era n-gram language model often takes gigabytes in memory, making them impractical to deploy on mobile devices. [89] Consequently, ASR systems from Google and Apple (as of 2017) deploy on servers and required a network connection to operate.[ citation needed ]

The first attempt at end-to-end ASR was the Connectionist Temporal Classification (CTC)-based system introduced by Alex Graves of Google DeepMind and Navdeep Jaitly of the University of Toronto in 2014. [90] The model consisted of RNNs and a CTC layer. Jointly, the RNN-CTC model learns the pronunciation and acoustic model together, however, it is incapable of learning the language model due to conditional independence assumptions, similar to an HMM. Consequently, CTC models can directly learn to map speech acoustics to English characters, but the models make many common spelling mistakes and must rely on a separate language model to finalize transcripts. Later, Baidu expanded on the work with extremely large datasets and demonstrated some commercial success in Mandarin and English. [91]

In 2016, the University of Oxford presented LipNet, [92] the first end-to-end sentence-level lipreading model, using spatiotemporal convolutions coupled with an RNN-CTC architecture, surpassing human-level performance in a restricted dataset. [93] A large-scale convolutional-RNN-CTC architecture was presented in 2018 by Google DeepMind, achieving 6 times better performance than human experts. [94] In 2019, Nvidia launched two CNN-CTC ASR models, Jasper and QuarzNet, with an overall performance word error rate (WER) of 3%. [95] [96] Similar to other deep learning applications, transfer learning and domain adaptation are important strategies for reusing and extending the capabilities of deep learning models, particularly due to the small size of available corpora in many languages and/or specific domains. [97] [98] [99]

In 2018, researchers at MIT Media Lab announced preliminary work on AlterEgo, a device that uses electrodes to read the neuromuscular signals users make as they subvocalize. [100] They trained a convolutional neural network to translate the electrode signals into words. [101]

Attention-based models

Attention-based ASR models were introduced by Chan et al. of Carnegie Mellon University and Google Brain, and Bahdanau et al. of the University of Montreal in 2016. [102] [103] The model named "Listen, Attend and Spell" (LAS), literally "listens" to the acoustic signal, pays "attention" to all parts of the signal and "spells" out the transcript one character at a time. Unlike CTC-based models, attention-based models require conditional-independence assumptions and can learn all the components of a speech recognizer directly. This means that during deployment, no a priori language model is required, making it less demanding for applications with limited memory.

Attention-based models immediately outperformed CTC models (with or without an external language model) and continued improving. [104] Latent Sequence Decomposition (LSD) was proposed by Carnegie Mellon University, MIT, and Google Brain to directly emit sub-word units that are more natural than English characters. [105] The University of Oxford and Google DeepMind extended LAS to "Watch, Listen, Attend and Spell" (WLAS) to handle lip reading and surpassed human-level performance. [106]

Applications

In-car systems

Voice commands may be used to initiate phone calls, select radio stations, or play music. Voice recognition capabilities vary across car make and model. Some models offer natural-language speech recognition, allowing the driver to use full sentences and common phrases in a conversational style. With such systems, fixed commands are not required.[ citation needed ]

Education

Automatic pronunciation assessment is the use of speech recognition to verify the correctness of speech, [107] as distinguished from assessment by a person. [108] Also called speech verification, pronunciation evaluation, and pronunciation scoring, the main application of this technology is computer-aided pronunciation teaching (CAPT) when combined with computer-aided instruction for computer-assisted language learning (CALL), speech remediation, or accent reduction. Pronunciation assessment does not determine unknown speech (as in dictation or automatic transcription) but instead, compares speech to a reference model for the words spoken, [109] [110] sometimes with inconsequential prosody such as intonation, pitch, tempo, rhythm, and stress. [111] Pronunciation assessment is also used in reading tutoring, for example in products such as Microsoft Teams [112] and Amira Learning. [113] Pronunciation assessment can also be used to help diagnose and treat speech disorders such as apraxia. [114]

Assessing intelligibility is essential for avoiding inaccuracies from accent bias, especially in high-stakes assessments, [115] [116] [117] from words with multiple correct pronunciations, [118] and from phoneme coding errors in digital pronunciation dictionaries. [119] In 2022, researchers found that some newer speech to text systems, based on end-to-end reinforcement learning to map audio signals directly into words, produce word and phrase confidence scores closely correlated with listener intelligibility. [120] In the Common European Framework of Reference for Languages (CEFR) assessment criteria for "overall phonological control", intelligibility outweighs formally correct pronunciation at all levels. [121]

Health care

Medical documentation

In the health care sector, speech recognition can be implemented in front-end or back-end medical documentation processes. In front-end speech recognition, the provider dictates into a speech-recognition engine, words are displayed as they are recognized, and the speaker is responsible for editing and signing off on the document. In back-end or deferred speech recognition the provider speaks into a digital dictation system, the voice is routed through a speech-recognition machine, and a draft document is routed along with the voice file to an editor, who edits/finalizes the draft and final report.[ citation needed ]

A major issue is that the American Recovery and Reinvestment Act of 2009 (ARRA) provides substantial financial benefits to physicians who utilize an Electronic Health Record (EHR) that complies with "Meaningful Use" standards. These standards require that substantial data be maintained by the EHR. The use of speech recognition is more naturally suited to the generation of narrative text, as part of a radiology/pathology interpretation, progress note or discharge summary; the ergonomic gains of using speech recognition to enter structured discrete data (e.g., numeric values or codes from a list or a controlled vocabulary) are relatively minimal for people who are sighted and who can operate a keyboard and mouse.

A more significant issue is that most EHRs have not been expressly tailored to take advantage of voice-recognition capabilities. A large part of a clinician's interaction with EHR involves navigation through the user interface that is heavily dependent on keyboard and mouse; voice-based navigation provides only modest ergonomic benefits. By contrast, many highly customized systems for radiology or pathology dictation implement voice "macros", where the use of certain phrases – e.g., "normal report", will automatically fill in a large number of default values and/or generate boilerplate, which vary with the type of exam – e.g., a chest X-ray vs. a gastrointestinal contrast series for a radiology system.

Therapeutic use

Prolonged use of speech recognition software in conjunction with word processors has shown benefits to short-term-memory restrengthening in brain AVM patients who have been treated with resection. Further research needs to be conducted to determine cognitive benefits for individuals whose AVMs have been treated using radiologic techniques.[ citation needed ]

Military

Aircraft

Substantial efforts have been devoted to the test and evaluation of speech recognition in fighter aircraft. Of particular note have been the US programme in speech recognition for the Advanced Fighter Technology Integration (AFTI)/F-16 aircraft (F-16 VISTA), the programme in France for Mirage aircraft, and UK programmes dealing with a variety of aircraft platforms. In these programmes, speech recognizers have been operated successfully, 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 the JAS-39 Gripen, Englund (2004) reported that recognition deteriorated with increasing g-loads. The study concluded that adaptation greatly improved the results in all cases and that the introduction of models for breathing was shown to improve recognition scores significantly. Contrary to what might have been expected, no effects of the broken English of the speakers were found. Spontaneous speech caused problems for the recognizer, as might have been expected. A restricted vocabulary, and above all, a proper syntax, could thus be expected to improve recognition accuracy substantially. [122]

The Eurofighter Typhoon employs a speaker-dependent system, requiring each pilot to create a template. The system is not used for safety-critical or weapon-critical tasks, such as weapon release or lowering of the undercarriage, but is used for many cockpit functions. Voice commands are confirmed by visual and/or aural feedback. The system is seen as a major benefit in the reduction of pilot workload, [123] and allows the pilot to assign targets with two voice commands or to a wingman with only five commands. [124]

Speaker-independent systems are under test for the F-35 Lightning II (JSF) and the Alenia Aermacchi M-346 Master lead-in fighter trainer. These systems have produced word accuracy scores in excess of 98%. [125] [ citation needed ]

Helicopters

The problems of achieving high recognition accuracy under stress and noise are particularly relevant in the helicopter environment as well as in the fighter environment. The acoustic noise problem is actually more severe in the helicopter environment, because of the high noise levels, and because helicopter pilots, in general, do not wear a facemask, which would reduce acoustic noise in the microphone. Substantial test and evaluation programmes, 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 included speech recognition in the Puma helicopter. Voice applications include control of communication radios, navigation systems, and an automated target handover system.

The overriding issue for voice 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.

Air traffic control

Training for air traffic controllers (ATC) represents an excellent application for speech recognition systems. Many ATC training systems currently require a trainer to act as a "pseudo-pilot", engaging in a voice dialog with the trainee controller, which simulates the dialog that the controller would have with real pilots. Speech recognition and synthesis techniques offer the potential to eliminate the need for a person to act as a pseudo-pilot, thus reducing training and support personnel.

In theory, air controller tasks are characterized by highly structured speech as the primary output, reducing the difficulty of the speech recognition task. In practice, this is rarely the case. 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 international ATC training organizations such as the Royal Australian Air Force and Civil Aviation Authorities in Italy, Brazil, and Canada use ATC simulators with speech recognition.[ citation needed ]

People with disabilities

Speech recognition programs can provide many benefit to those with disabilities. For individuals who are deaf or hard of hearing, speech recognition software can be used to generate captions of conversations. [126] Additionally, individuals who are blind (see blindness and education) or have poor vision can benefit from listening to textual content, as well as garner more functionality from a computer by issuing commands with their voice. [127]

The use of voice recognition software, in conjunction with a digital audio recorder and a personal computer running word-processing software, has proven useful for restoring damaged short-term memory capacity in individuals who have suffered a stroke or have undergone a craniotomy.[ citation needed ]

Speech recognition has proven very useful for those who have difficulty using their hands due to causes ranging from mild repetitive stress injuries to disabilities that preclude the use of conventional computer input devices. Individuals with physical disabilities can use voice commands and transcription to navigate electronics hands-free. [127] In fact, people who developed RSI from keyboard use became an early and urgent market for speech recognition. [128] [129] Speech recognition is used in deaf telephony, such as voicemail to text, relay services, and captioned telephone. Individuals with learning disabilities who struggle with thought-to-paper communication may benefit from the software, but the product's fallibility remains a significant consideration for many. [130] In addition, speech to text technology is only an effective aid for those with intellectual disabilities if the proper training and resources are provided (e.g. in the classroom setting). [131]

This type of technology can help those with dyslexia, but the potential benefits regarding other disabilities are still in question. Mistakes made by the software hinder its effectiveness, since misheard words take more time to fix. [132]

Other domains

ASR is now commonplace in the field of telephony. In telephony systems, ASR is predominantly used in contact centers by integrating it with IVR systems.

It is becoming more widespread in computer gaming and simulation.

Despite the high level of integration with word processing in general personal computing, in the field of document production, ASR has not seen the expected increases in use.

The improvement of mobile processor speeds has made speech recognition practical in smartphones. Speech is used mostly as a part of a user interface, for creating predefined or custom speech commands.

Performance

The performance of speech recognition systems is usually evaluated in terms of accuracy and speed. [137] [138] Accuracy is usually rated with word error rate (WER), whereas speed is measured in elapsed time. Other measures of accuracy include Single Word Error Rate (SWER) and Command Success Rate (CSR).

Speech recognition is complicated by many properties of speech. Vocalizations vary in terms of accent, pronunciation, articulation, roughness, dialect, nasality, pitch, volume, and speed. Speech is distorted by background noise, echoes, and recording characteristics. Accuracy of speech recognition may vary with the following: [139] [ citation needed ]

Accuracy

The accuracy of speech recognition may vary depending on the following factors:

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 confusing words (most notoriously, the E-set: "B, C, D, E, G, P, T, V, Z (when "Z" is pronounced "zee" rather than "zed", depending on region); an 8% error rate is considered good for this vocabulary. [140]
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, which is easier to recognize.
With discontinuous speech, full sentences separated by silence are used. The silence is easier to recognize similar to isolated speech.
With continuous speech naturally spoken sentences are used, which are harder to recognize.

Constraints are often represented by grammar.

Speech recognition is a multi-level pattern recognition task.

e.g., known word pronunciations or legal word sequences, which can compensate for errors or uncertainties at a lower level;

Sound is produced by air (or some other medium) vibration. Sound creates a wave that has two measures: amplitude (strength), and frequency (vibrations per second). Accuracy can be computed with the help of WER, which is calculated by aligning the recognized word and referenced word using dynamic string alignment. The problem may occur while computing the WER due to the difference between the sequence lengths of the recognized word and referenced word.

The formula to compute the word error rate (WER) is:

where s is the number of substitutions, d is the number of deletions, i is the number of insertions, and n is the number of word references.

While computing, the word recognition rate (WRR) is used. The formula is:

where h is the number of correctly recognized words:

Security

Speech recognition can become a means of attack, theft, or accidental operation. For example, activation words like "Alexa" spoken in an audio or video broadcast can cause devices in homes and offices to start listening for input inappropriately, or possibly take an unwanted action. [141] Voice-controlled devices may be accessible to unauthorized users. Attackers may be able to gain access to personal information, like calendars, address book contents, private messages, and documents. They may also be able to impersonate the user to send messages or make online purchases.

Two attacks have been demonstrated that use artificial sounds. One transmits ultrasound and attempts to send commands without people noticing. [142] The other adds small, inaudible distortions to other speech or music that are specially crafted to confuse the specific speech recognition system into recognizing music as speech, or to make what sounds like one command to a human sound like a different command to the system. [143]

Further information

Conferences

Regular conferences 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, include papers on speech processing.

Journal

The main journal is IEEE/ACM Transactions on Audio, Speech and Language Processing.

Books

Projects

The largest speech recognition-related project ongoing as of 2007 was the GALE project, which involves both speech recognition and translation components.

Software

See also

Lists

References

  1. ↑ "What Is Speech Recognition? | IBM". www.ibm.com. 28 September 2021. Retrieved 28 August 2025.
  2. ↑ P. Nguyen (2010). "Automatic classification of speaker characteristics". International Conference on Communications and Electronics 2010. pp. 147–152. doi:10.1109/ICCE.2010.5670700. ISBN   978-1-4244-7055-6. S2CID   13482115.
  3. ↑ "British English definition of voice recognition". Macmillan Publishers Limited. Archived from the original on 16 September 2011. Retrieved 21 February 2012.
  4. ↑ "voice recognition, definition of". WebFinance, Inc. Archived from the original on 3 December 2011. Retrieved 21 February 2012.
  5. ↑ "The Mailbag LG #114". Linuxgazette.net. Archived from the original on 19 February 2013. Retrieved 15 June 2013.
  6. ↑ Sarangi, Susanta; Sahidullah, Md; Saha, Goutam (September 2020). "Optimization of data-driven filterbank for automatic speaker verification". Digital Signal Processing. 104 102795. arXiv: 2007.10729 . Bibcode:2020DSP...10402795S. doi:10.1016/j.dsp.2020.102795. S2CID   220665533.
  7. ↑ Reynolds, Douglas; Rose, Richard (January 1995). "Robust text-independent speaker identification using Gaussian mixture speaker models" (PDF). IEEE Transactions on Speech and Audio Processing. 3 (1): 72–83. doi:10.1109/89.365379. ISSN   1063-6676. OCLC   26108901. S2CID   7319345. Archived (PDF) from the original on 8 March 2014. Retrieved 21 February 2014.
  8. ↑ "Speaker Identification (WhisperID)". Microsoft Research. Microsoft. Archived from the original on 25 February 2014. Retrieved 21 February 2014. When you speak to someone, they don't just recognize what you say: they recognize who you are. WhisperID will let computers do that, too, figuring out who you are by the way you sound.
  9. ↑ Alharbi, Sadeen; Alrazgan, Muna; Alrashed, Alanoud; Alnomasi, Turkiayh; Almojel, Raghad; Alharbi, Rimah; Alharbi, Saja; Alturki, Sahar; Alshehri, Fatimah; Almojil, Maha (2021). "Automatic Speech Recognition: Systematic Literature Review". IEEE Access. 9: 131858–131876. Bibcode:2021IEEEA...9m1858A. doi: 10.1109/ACCESS.2021.3112535 . ISSN   2169-3536.
  10. ↑ Li, Suo; You, Jinchi; Zhang, Xin (August 2022). "Overview and Analysis of Speech Recognition". 2022 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA). pp. 391–395. doi:10.1109/AEECA55500.2022.9919050. ISBN   978-1-6654-8090-1.
  11. ↑ "Obituaries: Stephen Balashek". The Star-Ledger. 22 July 2012. Archived from the original on 4 April 2019. Retrieved 9 September 2024.
  12. ↑ "IBM-Shoebox-front.jpg". androidauthority.net. Archived from the original on 9 August 2018. Retrieved 4 April 2019.
  13. ↑ Juang, B. H.; Rabiner, Lawrence R. "Automatic speech recognition–a brief history of the technology development" (PDF). p. 6. Archived (PDF) from the original on 17 August 2014. Retrieved 17 January 2015.
  14. 1 2 Melanie Pinola (2 November 2011). "Speech Recognition Through the Decades: How We Ended Up With Siri". PC World. Archived from the original on 3 November 2018. Retrieved 22 October 2018.
  15. ↑ Gray, Robert M. (2010). "A History of Realtime Digital Speech on Packet Networks: Part II of Linear Predictive Coding and the Internet Protocol" (PDF). Found. Trends Signal Process. 3 (4): 203–303. doi: 10.1561/2000000036 . ISSN   1932-8346. Archived (PDF) from the original on 9 October 2022. Retrieved 9 September 2024.
  16. ↑ John R. Pierce (1969). "Whither speech recognition?". Journal of the Acoustical Society of America. 46 (48): 1049–1051. Bibcode:1969ASAJ...46.1049P. doi:10.1121/1.1911801.
  17. ↑ Benesty, Jacob; Sondhi, M. M.; Huang, Yiteng (2008). Springer Handbook of Speech Processing. Springer Science & Business Media. ISBN   978-3540491255.
  18. ↑ John Makhoul. "ISCA Medalist: For leadership and extensive contributions to speech and language processing". Archived from the original on 24 January 2018. Retrieved 23 January 2018.
  19. ↑ Blechman, R. O.; Blechman, Nicholas (23 June 2008). "Hello, Hal". The New Yorker. Archived from the original on 20 January 2015. Retrieved 17 January 2015.
  20. ↑ Klatt, Dennis H. (1977). "Review of the ARPA speech understanding project". The Journal of the Acoustical Society of America. 62 (6): 1345–1366. Bibcode:1977ASAJ...62.1345K. doi:10.1121/1.381666.
  21. ↑ Rabiner (1984). "The Acoustics, Speech, and Signal Processing Society. A Historical Perspective" (PDF). Archived (PDF) from the original on 9 August 2017. Retrieved 23 January 2018.
  22. ↑ "First-Hand:The Hidden Markov Model – Engineering and Technology History Wiki". ethw.org. 12 January 2015. Archived from the original on 3 April 2018. Retrieved 1 May 2018.
  23. 1 2 "James Baker interview". Archived from the original on 28 August 2017. Retrieved 9 February 2017.
  24. ↑ "Pioneering Speech Recognition". 7 March 2012. Archived from the original on 19 February 2015. Retrieved 18 January 2015.
  25. ↑ Huang, Xuedong; Baker, James; Reddy, Raj (January 2014). "A historical perspective of speech recognition". Communications of the ACM. 57 (1): 94–103. doi:10.1145/2500887. ISSN   0001-0782. S2CID   6175701. Archived from the original on 8 December 2023.
  26. ↑ Juang, B. H.; Rabiner, Lawrence R. Automatic speech recognition–a brief history of the technology development (PDF) (Report). p. 10. Archived (PDF) from the original on 17 August 2014. Retrieved 17 January 2015.
  27. ↑ Li, Xiaochang (1 July 2023). ""There's No Data Like More Data": Automatic Speech Recognition and the Making of Algorithmic Culture" . Osiris. 38: 165–182. doi:10.1086/725132. ISSN   0369-7827. S2CID   259502346.
  28. ↑ "History of Speech Recognition". Dragon Medical Transcription. Archived from the original on 13 August 2015. Retrieved 17 January 2015.
  29. ↑ Billi, Roberto; Canavesio, Franco; Ciaramella, Alberto; Nebbia, Luciano (1 November 1995). "Interactive voice technology at work: The CSELT experience" . Speech Communication. 17 (3): 263–271. doi:10.1016/0167-6393(95)00030-R.
  30. 1 2 Xuedong Huang; James Baker; Raj Reddy (January 2014). "A Historical Perspective of Speech Recognition". Communications of the ACM. Archived from the original on 20 January 2015. Retrieved 20 January 2015.
  31. ↑ Kevin McKean (8 April 1980). "When Cole talks, computers listen". Sarasota Journal. AP. Retrieved 23 November 2015.
  32. ↑ "ACT/Apricot - Apricot history". actapricot.org. Archived from the original on 21 December 2016. Retrieved 2 February 2016.
  33. ↑ Melanie Pinola (2 November 2011). "Speech Recognition Through the Decades: How We Ended Up With Siri". PC World. Archived from the original on 13 January 2017. Retrieved 28 July 2017.
  34. ↑ "Ray Kurzweil biography". KurzweilAINetwork. Archived from the original on 5 February 2014. Retrieved 25 September 2014.
  35. ↑ Juang, B.H.; Rabiner, Lawrence. Automatic Speech Recognition – A Brief History of the Technology Development (PDF) (Report). Archived (PDF) from the original on 9 August 2017. Retrieved 28 July 2017.
  36. ↑ "Nuance Exec on iPhone 4S, Siri, and the Future of Speech". Tech.pinions. 10 October 2011. Archived from the original on 19 November 2011. Retrieved 23 November 2011.
  37. ↑ "Switchboard-1 Release 2". Archived from the original on 11 July 2017. Retrieved 26 July 2017.
  38. ↑ Jason Kincaid (13 February 2011). "The Power of Voice: A Conversation With The Head Of Google's Speech Technology". Tech Crunch. Archived from the original on 21 July 2015. Retrieved 21 July 2015.
  39. ↑ Froomkin, Dan (5 May 2015). "THE COMPUTERS ARE LISTENING". The Intercept. Archived from the original on 27 June 2015. Retrieved 20 June 2015.
  40. ↑ Herve Bourlard and Nelson Morgan, Connectionist Speech Recognition: A Hybrid Approach, The Kluwer International Series in Engineering and Computer Science; v. 247, Boston: Kluwer Academic Publishers, 1994.
  41. 1 2 Sepp Hochreiter; J. Schmidhuber (1997). "Long Short-Term Memory". Neural Computation. 9 (8): 1735–1780. doi:10.1162/neco.1997.9.8.1735. PMID   9377276. S2CID   1915014.
  42. ↑ Schmidhuber, Jürgen (2015). "Deep learning in neural networks: An overview". Neural Networks. 61: 85–117. arXiv: 1404.7828 . doi:10.1016/j.neunet.2014.09.003. PMID   25462637. S2CID   11715509.
  43. ↑ Alex Graves, Santiago Fernandez, Faustino Gomez, and Jürgen Schmidhuber (2006). Connectionist temporal classification: Labelling unsegmented sequence data with recurrent neural nets Archived 9 September 2024 at the Wayback Machine . Proceedings of ICML'06, pp. 369–376.
  44. ↑ Santiago Fernandez, Alex Graves, and Jürgen Schmidhuber (2007). An application of recurrent neural networks to discriminative keyword spotting [ permanent dead link ]. Proceedings of ICANN (2), pp. 220–229.
  45. 1 2 Haşim Sak, Andrew Senior, Kanishka Rao, Françoise Beaufays and Johan Schalkwyk (September 2015): " "Google voice search: faster and more accurate". Archived from the original on 9 March 2016. Retrieved 5 April 2016.."
  46. ↑ Dosovitskiy, Alexey; Beyer, Lucas; Kolesnikov, Alexander; Weissenborn, Dirk; Zhai, Xiaohua; Unterthiner, Thomas; Dehghani, Mostafa; Minderer, Matthias; Heigold, Georg; Gelly, Sylvain; Uszkoreit, Jakob; Houlsby, Neil (3 June 2021). "An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale". arXiv: 2010.11929 [cs.CV].
  47. ↑ Wu, Haiping; Xiao, Bin; Codella, Noel; Liu, Mengchen; Dai, Xiyang; Yuan, Lu; Zhang, Lei (29 March 2021). "CvT: Introducing Convolutions to Vision Transformers". arXiv: 2103.15808 [cs.CV].
  48. ↑ Vaswani, Ashish; Shazeer, Noam; Parmar, Niki; Uszkoreit, Jakob; Jones, Llion; Gomez, Aidan N; Kaiser, Łukasz; Polosukhin, Illia (2017). "Attention is All you Need". Advances in Neural Information Processing Systems. 30. Curran Associates. Archived from the original on 9 September 2024. Retrieved 9 September 2024.
  49. ↑ Devlin, Jacob; Chang, Ming-Wei; Lee, Kenton; Toutanova, Kristina (24 May 2019). "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding". arXiv: 1810.04805 [cs.CL].
  50. 1 2 Gong, Yuan; Chung, Yu-An; Glass, James (8 July 2021). "AST: Audio Spectrogram Transformer". arXiv: 2104.01778 [cs.SD].
  51. 1 2 Ristea, Nicolae-Catalin; Ionescu, Radu Tudor; Khan, Fahad Shahbaz (20 June 2022). "SepTr: Separable Transformer for Audio Spectrogram Processing". arXiv: 2203.09581 [cs.CV].
  52. 1 2 Lohrenz, Timo; Li, Zhengyang; Fingscheidt, Tim (14 July 2021). "Multi-Encoder Learning and Stream Fusion for Transformer-Based End-to-End Automatic Speech Recognition". arXiv: 2104.00120 [eess.AS].
  53. ↑ "Li Deng". Li Deng Site. Archived from the original on 9 September 2024. Retrieved 9 September 2024.
  54. ↑ NIPS Workshop: Deep Learning for Speech Recognition and Related Applications, Whistler, BC, Canada, Dec. 2009 (Organizers: Li Deng, Geoff Hinton, D. Yu).
  55. 1 2 3 Hinton, Geoffrey; Deng, Li; Yu, Dong; Dahl, George; Mohamed, Abdel-Rahman; Jaitly, Navdeep; Senior, Andrew; Vanhoucke, Vincent; Nguyen, Patrick; Sainath, Tara; Kingsbury, Brian (2012). "Deep Neural Networks for Acoustic Modeling in Speech Recognition: The shared views of four research groups". IEEE Signal Processing Magazine. 29 (6): 82–97. Bibcode:2012ISPM...29...82H. doi:10.1109/MSP.2012.2205597. S2CID   206485943.
  56. 1 2 Deng, L.; Hinton, G.; Kingsbury, B. (2013). "New types of deep neural network learning for speech recognition and related applications: An overview". 2013 IEEE International Conference on Acoustics, Speech and Signal Processing: New types of deep neural network learning for speech recognition and related applications: An overview. p. 8599. doi:10.1109/ICASSP.2013.6639344. ISBN   978-1-4799-0356-6. S2CID   13953660.
  57. 1 2 Markoff, John (23 November 2012). "Scientists See Promise in Deep-Learning Programs". New York Times. Archived from the original on 30 November 2012. Retrieved 20 January 2015.
  58. ↑ Morgan, Bourlard, Renals, Cohen, Franco (1993) "Hybrid neural network/hidden Markov model systems for continuous speech recognition. ICASSP/IJPRAI"
  59. ↑ T. Robinson (1992). "A real-time recurrent error propagation network word recognition system". [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing. pp. 617–620 vol.1. doi:10.1109/ICASSP.1992.225833. ISBN   0-7803-0532-9. S2CID   62446313.
  60. ↑ Waibel, Hanazawa, Hinton, Shikano, Lang. (1989) "Phoneme recognition using time-delay neural networks Archived 25 February 2021 at the Wayback Machine . IEEE Transactions on Acoustics, Speech, and Signal Processing."
  61. ↑ Baker, J.; Li Deng; Glass, J.; Khudanpur, S.; Chin-Hui Lee; Morgan, N.; O'Shaughnessy, D. (2009). "Developments and Directions in Speech Recognition and Understanding, Part 1". IEEE Signal Processing Magazine. 26 (3): 75–80. Bibcode:2009ISPM...26...75B. doi:10.1109/MSP.2009.932166. hdl: 1721.1/51891 . S2CID   357467.
  62. ↑ Sepp Hochreiter (1991), Untersuchungen zu dynamischen neuronalen Netzen Archived 6 March 2015 at the Wayback Machine , Diploma thesis. Institut f. Informatik, Technische Univ. Munich. Advisor: J. Schmidhuber.
  63. ↑ Bengio, Y. (1991). Artificial Neural Networks and their Application to Speech/Sequence Recognition (Ph.D. thesis). McGill University.
  64. ↑ Deng, L.; Hassanein, K.; Elmasry, M. (1994). "Analysis of the correlation structure for a neural predictive model with application to speech recognition". Neural Networks. 7 (2): 331–339. doi:10.1016/0893-6080(94)90027-2.
  65. ↑ Keynote talk: Recent Developments in Deep Neural Networks. ICASSP, 2013 (by Geoff Hinton).
  66. 1 2 Keynote talk: "Achievements and Challenges of Deep Learning: From Speech Analysis and Recognition To Language and Multimodal Processing Archived 5 March 2021 at the Wayback Machine ," Interspeech, September 2014 (by Li Deng).
  67. ↑ "Improvements in voice recognition software increase". TechRepublic.com. 27 August 2002. Archived from the original on 23 October 2018. Retrieved 22 October 2018. Maners said IBM has worked on advancing speech recognition ... or on the floor of a noisy trade show.
  68. ↑ "Voice Recognition To Ease Travel Bookings: Business Travel News". BusinessTravelNews.com. 3 March 1997. Archived from the original on 9 September 2024. Retrieved 9 September 2024. The earliest applications of speech recognition software were dictation ... Four months ago, IBM introduced a 'continual dictation product' designed to ... debuted at the National Business Travel Association trade show in 1994.
  69. ↑ Ellis Booker (14 March 1994). "Voice recognition enters the mainstream". Computerworld . p. 45. Just a few years ago, speech recognition was limited to ...
  70. ↑ "Microsoft researchers achieve new conversational speech recognition milestone". Microsoft . 21 August 2017. Archived from the original on 9 September 2024. Retrieved 9 September 2024.
  71. ↑ Waibel, A.; Hanazawa, T.; Hinton, G.; Shikano, K.; Lang, K. J. (1989). "Phoneme recognition using time-delay neural networks". IEEE Transactions on Acoustics, Speech, and Signal Processing. 37 (3): 328–339. doi:10.1109/29.21701. hdl: 10338.dmlcz/135496 . S2CID   9563026.
  72. ↑ Bird, Jordan J.; Wanner, Elizabeth; Ekárt, Anikó; Faria, Diego R. (2020). "Optimisation of phonetic aware speech recognition through multi-objective evolutionary algorithms" (PDF). Expert Systems with Applications. 153 113402. Elsevier BV. doi:10.1016/j.eswa.2020.113402. ISSN   0957-4174. S2CID   216472225. Archived (PDF) from the original on 9 September 2024. Retrieved 9 September 2024.
  73. ↑ Wu, J.; Chan, C. (1993). "Isolated Word Recognition by Neural Network Models with Cross-Correlation Coefficients for Speech Dynamics". IEEE Transactions on Pattern Analysis and Machine Intelligence. 15 (11): 1174–1185. doi:10.1109/34.244678.
  74. ↑ S. A. Zahorian, A. M. Zimmer, and F. Meng, (2002) "Vowel Classification for Computer based Visual Feedback for Speech Training for the Hearing Impaired," in ICSLP 2002
  75. ↑ Hu, Hongbing; Zahorian, Stephen A. (2010). "Dimensionality Reduction Methods for HMM Phonetic Recognition" (PDF). ICASSP 2010. Archived (PDF) from the original on 6 July 2012.
  76. ↑ Fernandez, Santiago; Graves, Alex; Schmidhuber, Jürgen (2007). "Sequence labelling in structured domains with hierarchical recurrent neural networks" (PDF). Proceedings of IJCAI. Archived (PDF) from the original on 15 August 2017.
  77. ↑ Graves, Alex; Mohamed, Abdel-rahman; Hinton, Geoffrey (2013). "Speech recognition with deep recurrent neural networks". arXiv: 1303.5778 [cs.NE]. ICASSP 2013.
  78. ↑ Waibel, Alex (1989). "Modular Construction of Time-Delay Neural Networks for Speech Recognition" (PDF). Neural Computation. 1 (1): 39–46. doi:10.1162/neco.1989.1.1.39. S2CID   236321. Archived (PDF) from the original on 29 June 2016.
  79. ↑ Maas, Andrew L.; Le, Quoc V.; O'Neil, Tyler M.; Vinyals, Oriol; Nguyen, Patrick; Ng, Andrew Y. (2012). "Recurrent Neural Networks for Noise Reduction in Robust ASR". Proceedings of Interspeech 2012.
  80. 1 2 Deng, Li; Yu, Dong (2014). "Deep Learning: Methods and Applications" (PDF). Foundations and Trends in Signal Processing. 7 (3–4): 197–387. CiteSeerX   10.1.1.691.3679 . doi:10.1561/2000000039. Archived (PDF) from the original on 22 October 2014.
  81. ↑ Yu, D.; Deng, L.; Dahl, G. (2010). "Roles of Pre-Training and Fine-Tuning in Context-Dependent DBN-HMMs for Real-World Speech Recognition" (PDF). NIPS Workshop on Deep Learning and Unsupervised Feature Learning.
  82. ↑ Dahl, George E.; Yu, Dong; Deng, Li; Acero, Alex (2012). "Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition". IEEE Transactions on Audio, Speech, and Language Processing. 20 (1): 30–42. Bibcode:2012ITASL..20...30D. doi:10.1109/TASL.2011.2134090. S2CID   14862572.
  83. ↑ Deng L., Li, J., Huang, J., Yao, K., Yu, D., Seide, F. et al. Recent Advances in Deep Learning for Speech Research at Microsoft Archived 9 September 2024 at the Wayback Machine . ICASSP, 2013.
  84. 1 2 Yu, D.; Deng, L. (2014). "Automatic Speech Recognition: A Deep Learning Approach (Publisher: Springer)".{{cite journal}}: Cite journal requires |journal= (help)
  85. ↑ Deng, L.; Li, Xiao (2013). "Machine Learning Paradigms for Speech Recognition: An Overview" (PDF). IEEE Transactions on Audio, Speech, and Language Processing. 21 (5): 1060–1089. Bibcode:2013ITASL..21.1060D. doi:10.1109/TASL.2013.2244083. S2CID   16585863. Archived (PDF) from the original on 9 September 2024. Retrieved 9 September 2024.
  86. ↑ Schmidhuber, Jürgen (2015). "Deep Learning". Scholarpedia. 10 (11): 32832. Bibcode:2015SchpJ..1032832S. doi: 10.4249/scholarpedia.32832 .{{cite journal}}: CS1 maint: article number as page number (link)
  87. ↑ L. Deng, M. Seltzer, D. Yu, A. Acero, A. Mohamed, and G. Hinton (2010) Binary Coding of Speech Spectrograms Using a Deep Auto-encoder. Interspeech.
  88. ↑ Tüske, Zoltán; Golik, Pavel; Schlüter, Ralf; Ney, Hermann (2014). "Acoustic Modeling with Deep Neural Networks Using Raw Time Signal for LVCSR" (PDF). Interspeech 2014. Archived (PDF) from the original on 21 December 2016.
  89. ↑ Jurafsky, Daniel (2016). Speech and Language Processing.
  90. ↑ Graves, Alex (2014). "Towards End-to-End Speech Recognition with Recurrent Neural Networks" (PDF). ICML. Archived from the original (PDF) on 10 January 2017. Retrieved 22 July 2019.
  91. ↑ Amodei, Dario (2016). "Deep Speech 2: End-to-End Speech Recognition in English and Mandarin". arXiv: 1512.02595 [cs.CL].
  92. ↑ "LipNet: How easy do you think lipreading is?". YouTube. 4 November 2016. Archived from the original on 27 April 2017. Retrieved 5 May 2017.
  93. ↑ Assael, Yannis; Shillingford, Brendan; Whiteson, Shimon; de Freitas, Nando (5 November 2016). "LipNet: End-to-End Sentence-level Lipreading". arXiv: 1611.01599 [cs.CV].
  94. ↑ Shillingford, Brendan; Assael, Yannis; Hoffman, Matthew W.; Paine, Thomas; Hughes, Cían; Prabhu, Utsav; Liao, Hank; Sak, Hasim; Rao, Kanishka (13 July 2018). "Large-Scale Visual Speech Recognition". arXiv: 1807.05162 [cs.CV].
  95. ↑ Li, Jason; Lavrukhin, Vitaly; Ginsburg, Boris; Leary, Ryan; Kuchaiev, Oleksii; Cohen, Jonathan M.; Nguyen, Huyen; Gadde, Ravi Teja (2019). "Jasper: An End-to-End Convolutional Neural Acoustic Model". Interspeech 2019. pp. 71–75. arXiv: 1904.03288 . doi:10.21437/Interspeech.2019-1819.
  96. ↑ Kriman, Samuel; Beliaev, Stanislav; Ginsburg, Boris; Huang, Jocelyn; Kuchaiev, Oleksii; Lavrukhin, Vitaly; Leary, Ryan; Li, Jason; Zhang, Yang (22 October 2019), QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions, arXiv: 1910.10261
  97. ↑ Medeiros, Eduardo; Corado, Leonel; Rato, Luís; Quaresma, Paulo; Salgueiro, Pedro (May 2023). "Domain Adaptation Speech-to-Text for Low-Resource European Portuguese Using Deep Learning". Future Internet. 15 (5): 159. doi: 10.3390/fi15050159 . ISSN   1999-5903.
  98. ↑ Joshi, Raviraj; Singh, Anupam (May 2022). Malmasi, Shervin; Rokhlenko, Oleg; Ueffing, Nicola; Guy, Ido; Agichtein, Eugene; Kallumadi, Surya (eds.). "A Simple Baseline for Domain Adaptation in End to End ASR Systems Using Synthetic Data". Proceedings of the Fifth Workshop on E-Commerce and NLP (ECNLP 5). Dublin, Ireland: Association for Computational Linguistics: 244–249. arXiv: 2206.13240 . doi:10.18653/v1/2022.ecnlp-1.28.
  99. ↑ Sukhadia, Vrunda N.; Umesh, S. (9 January 2023). "Domain Adaptation of Low-Resource Target-Domain Models Using Well-Trained ASR Conformer Models". 2022 IEEE Spoken Language Technology Workshop (SLT). IEEE. pp. 295–301. arXiv: 2202.09167 . doi:10.1109/SLT54892.2023.10023233. ISBN   979-8-3503-9690-4.
  100. ↑ Petrova, Magdalena (10 April 2018). "MIT developed a headset that gives a voice to the voice inside your head". CNBC. Retrieved 11 September 2025.
  101. ↑ Kapur, Arnav; Kapur, Shreyas; Maes, Pattie (2018). "AlterEgo: A Personalized Wearable Silent Speech Interface". IUI '18: Proceedings of the 23rd International Conference on Intelligent User Interfaces: 43–53.
  102. ↑ Chan, William; Jaitly, Navdeep; Le, Quoc; Vinyals, Oriol (2016). "Listen, Attend and Spell: A Neural Network for Large Vocabulary Conversational Speech Recognition" (PDF). ICASSP. Archived (PDF) from the original on 9 September 2024. Retrieved 9 September 2024.
  103. ↑ Bahdanau, Dzmitry (2016). "End-to-End Attention-based Large Vocabulary Speech Recognition". arXiv: 1508.04395 [cs.CL].
  104. ↑ Chorowski, Jan; Jaitly, Navdeep (8 December 2016). "Towards better decoding and language model integration in sequence to sequence models". arXiv: 1612.02695 [cs.NE].
  105. ↑ Chan, William; Zhang, Yu; Le, Quoc; Jaitly, Navdeep (10 October 2016). "Latent Sequence Decompositions". arXiv: 1610.03035 [stat.ML].
  106. ↑ Chung, Joon Son; Senior, Andrew; Vinyals, Oriol; Zisserman, Andrew (16 November 2016). "Lip Reading Sentences in the Wild". 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 3444–3453. arXiv: 1611.05358 . doi:10.1109/CVPR.2017.367. ISBN   978-1-5386-0457-1. S2CID   1662180.
  107. ↑ El Kheir, Yassine; et al. (21 October 2023), Automatic Pronunciation Assessment — A Review, Conference on Empirical Methods in Natural Language Processing, arXiv: 2310.13974 , S2CID   264426545
  108. ↑ Isaacs, Talia; Harding, Luke (July 2017). "Pronunciation assessment". Language Teaching. 50 (3): 347–366. doi: 10.1017/S0261444817000118 . ISSN   0261-4448. S2CID   209353525.
  109. ↑ Loukina, Anastassia; et al. (6 September 2015), "Pronunciation accuracy and intelligibility of non-native speech" (PDF), INTERSPEECH 2015, Dresden, Germany: International Speech Communication Association, pp. 1917–1921, archived (PDF) from the original on 9 September 2024, retrieved 9 September 2024, only 16% of the variability in word-level intelligibility can be explained by the presence of obvious mispronunciations.
  110. ↑ O'Brien, Mary Grantham; et al. (31 December 2018). "Directions for the future of technology in pronunciation research and teaching". Journal of Second Language Pronunciation. 4 (2): 182–207. doi: 10.1075/jslp.17001.obr . hdl: 2066/199273 . ISSN   2215-1931. S2CID   86440885. pronunciation researchers are primarily interested in improving L2 learners' intelligibility and comprehensibility, but they have not yet collected sufficient amounts of representative and reliable data (speech recordings with corresponding annotations and judgments) indicating which errors affect these speech dimensions and which do not. These data are essential to train ASR algorithms to assess L2 learners' intelligibility.
  111. ↑ Eskenazi, Maxine (January 1999). "Using automatic speech processing for foreign language pronunciation tutoring: Some issues and a prototype". Language Learning & Technology. 2 (2): 62–76. doi: 10.64152/10125/25043 . Archived from the original on 9 September 2024. Retrieved 11 February 2023.
  112. ↑ Tholfsen, Mike (9 February 2023). "Reading Coach in Immersive Reader plus new features coming to Reading Progress in Microsoft Teams". Techcommunity Education Blog. Microsoft. Archived from the original on 9 September 2024. Retrieved 12 February 2023.
  113. ↑ Banerji, Olina (7 March 2023). "Schools Are Using Voice Technology to Teach Reading. Is It Helping?". EdSurge News. Archived from the original on 9 September 2024. Retrieved 7 March 2023.
  114. ↑ Hair, Adam; et al. (19 June 2018). "Apraxia world: A speech therapy game for children with speech sound disorders". Proceedings of the 17th ACM Conference on Interaction Design and Children (PDF). pp. 119–131. doi:10.1145/3202185.3202733. ISBN   9781450351522. S2CID   13790002. Archived (PDF) from the original on 9 September 2024. Retrieved 9 September 2024.
  115. ↑ "Computer says no: Irish vet fails oral English test needed to stay in Australia". The Guardian. Australian Associated Press. 8 August 2017. Archived from the original on 9 September 2024. Retrieved 12 February 2023.
  116. ↑ Ferrier, Tracey (9 August 2017). "Australian ex-news reader with English degree fails robot's English test". The Sydney Morning Herald. Archived from the original on 9 September 2024. Retrieved 12 February 2023.
  117. ↑ Main, Ed; Watson, Richard (9 February 2022). "The English test that ruined thousands of lives". BBC News. Archived from the original on 9 September 2024. Retrieved 12 February 2023.
  118. ↑ Joyce, Katy Spratte (24 January 2023). "13 Words That Can Be Pronounced Two Ways". Reader's Digest. Archived from the original on 9 September 2024. Retrieved 23 February 2023.
  119. ↑ E.g., CMUDICT, "The CMU Pronouncing Dictionary". www.speech.cs.cmu.edu. Archived from the original on 15 August 2010. Retrieved 15 February 2023. Compare "four" given as "F AO R" with the vowel AO as in "caught," to "row" given as "R OW" with the vowel OW as in "oat."
  120. ↑ Tu, Zehai; Ma, Ning; Barker, Jon (2022). "Unsupervised Uncertainty Measures of Automatic Speech Recognition for Non-intrusive Speech Intelligibility Prediction" (PDF). Proc. Interspeech 2022. INTERSPEECH 2022. ISCA. pp. 3493–3497. doi:10.21437/Interspeech.2022-10408. Archived (PDF) from the original on 9 September 2024. Retrieved 17 December 2023.
  121. ↑ Common European framework of reference for languages learning, teaching, assessment: Companion volume with new descriptors. Language Policy Programme, Education Policy Division, Education Department, Council of Europe. February 2018. p. 136. OCLC   1090351600. Archived from the original on 9 September 2024. Retrieved 9 September 2024.
  122. ↑ Englund, Christine (2004). Speech recognition in the JAS 39 Gripen aircraft: Adaptation to speech at different G-loads (PDF) (Masters thesis thesis). Stockholm Royal Institute of Technology. Archived (PDF) from the original on 2 October 2008.
  123. ↑ "The Cockpit". Eurofighter Typhoon. Archived from the original on 1 March 2017.
  124. ↑ "Eurofighter Typhoon – The world's most advanced fighter aircraft". www.eurofighter.com. Archived from the original on 11 May 2013. Retrieved 1 May 2018.
  125. ↑ Schutte, John (15 October 2007). "Researchers fine-tune F-35 pilot-aircraft speech system". United States Air Force. Archived from the original on 20 October 2007.
  126. ↑ "Overcoming Communication Barriers in the Classroom". MassMATCH. 18 March 2010. Archived from the original on 25 July 2013. Retrieved 15 June 2013.
  127. 1 2 "Speech Recognition for Learning". National Center for Technology Innovation. 2010. Archived from the original on 13 April 2014. Retrieved 26 March 2014.
  128. ↑ "Speech recognition for disabled people". Archived from the original on 4 April 2008.
  129. ↑ Friends International Support Group
  130. ↑ Garrett, Jennifer Tumlin; et al. (2011). "Using Speech Recognition Software to Increase Writing Fluency for Individuals with Physical Disabilities". Journal of Special Education Technology. 26 (1): 25–41. doi:10.1177/016264341102600104. S2CID   142730664. Archived from the original on 9 September 2024. Retrieved 9 September 2024.
  131. ↑ Forgrave, Karen E. "Assistive Technology: Empowering Students with Disabilities." Clearing House 75.3 (2002): 122–6. Web.
  132. ↑ Tang, K. W.; Kamoua, Ridha; Sutan, Victor (2004). "Speech Recognition Technology for Disabilities Education". Journal of Educational Technology Systems. 33 (2): 173–84. CiteSeerX   10.1.1.631.3736 . doi:10.2190/K6K8-78K2-59Y7-R9R2. S2CID   143159997.
  133. ↑ "Projects: Planetary Microphones". The Planetary Society. Archived from the original on 27 January 2012.
  134. ↑ Caridakis, George; Castellano, Ginevra; Kessous, Loic; Raouzaiou, Amaryllis; Malatesta, Lori; Asteriadis, Stelios; Karpouzis, Kostas (19 September 2007). "Multimodal emotion recognition from expressive faces, body gestures and speech". Artificial Intelligence and Innovations 2007: From Theory to Applications. IFIP the International Federation for Information Processing. Vol. 247. Springer US. pp. 375–388. doi:10.1007/978-0-387-74161-1_41. ISBN   978-0-387-74160-4.
  135. ↑ "What is real-time captioning? | DO-IT". www.washington.edu. Archived from the original on 9 September 2024. Retrieved 11 April 2021.
  136. ↑ Zheng, Thomas Fang; Li, Lantian (2017). Robustness-Related Issues in Speaker Recognition. SpringerBriefs in Electrical and Computer Engineering. Singapore: Springer Singapore. doi:10.1007/978-981-10-3238-7. ISBN   978-981-10-3237-0. Archived from the original on 9 September 2024. Retrieved 9 September 2024.
  137. ↑ Ciaramella, Alberto. "A prototype performance evaluation report." Sundial workpackage 8000 (1993).
  138. ↑ Gerbino, E.; Baggia, P.; Ciaramella, A.; Rullent, C. (1993). "Test and evaluation of a spoken dialogue system". IEEE International Conference on Acoustics Speech and Signal Processing. pp. 135–138 vol.2. doi:10.1109/ICASSP.1993.319250. ISBN   0-7803-0946-4. S2CID   57374050.
  139. ↑ National Institute of Standards and Technology. "The History of Automatic Speech Recognition Evaluation at NIST Archived 8 October 2013 at the Wayback Machine ".
  140. ↑ "Letter Names Can Cause Confusion and Other Things to Know About Letter–Sound Relationships". NAEYC. Archived from the original on 9 September 2024. Retrieved 27 October 2023.
  141. ↑ "Listen Up: Your AI Assistant Goes Crazy For NPR Too". NPR . 6 March 2016. Archived from the original on 23 July 2017.
  142. ↑ Claburn, Thomas (25 August 2017). "Is it possible to control Amazon Alexa, Google Now using inaudible commands? Absolutely". The Register . Archived from the original on 2 September 2017.
  143. ↑ "Attack Targets Automatic Speech Recognition Systems". vice.com. 31 January 2018. Archived from the original on 3 March 2018. Retrieved 1 May 2018.
  144. ↑ Beigi, Homayoon (2011). Fundamentals of Speaker Recognition. New York: Springer. ISBN   978-0-387-77591-3. Archived from the original on 31 January 2018.
  145. ↑ Povey, D., Ghoshal, A., Boulianne, G., Burget, L., Glembek, O., Goel, N., ... & Vesely, K. (2011). The Kaldi speech recognition toolkit. In IEEE 2011 workshop on automatic speech recognition and understanding (No. CONF). IEEE Signal Processing Society.
  146. ↑ "Common Voice by Mozilla". voice.mozilla.org. Archived from the original on 27 February 2020. Retrieved 9 November 2019.
  147. ↑ "A TensorFlow implementation of Baidu's DeepSpeech architecture: mozilla/DeepSpeech". 9 November 2019. Archived from the original on 9 September 2024. Retrieved 9 September 2024– via GitHub.
  148. ↑ "GitHub - tensorflow/docs: TensorFlow documentation". 9 November 2019. Archived from the original on 9 September 2024. Retrieved 9 September 2024– via GitHub.
  149. ↑ "Coqui, a startup providing open speech tech for everyone". GitHub. Archived from the original on 9 September 2024. Retrieved 7 March 2022.
  150. ↑ Coffey, Donavyn (28 April 2021). "Māori are trying to save their language from Big Tech". Wired UK. ISSN   1357-0978. Archived from the original on 9 September 2024. Retrieved 16 October 2021.
  151. ↑ "Why you should move from DeepSpeech to coqui.ai". Mozilla Discourse. 7 July 2021. Retrieved 16 October 2021.
  152. ↑ "Type with your voice". Archived from the original on 9 September 2024. Retrieved 9 September 2024.
  153. ↑ "Use voice recognition in Windows". Archived from the original on 9 April 2025.

Further reading

This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.