Deep learningDeep learning is part of a broader family of machine learning methods, which is based on artificial neural networks with representation learning. The adjective "deep" in deep learning refers to the use of multiple layers in the network. Methods used can be either supervised, semi-supervised or unsupervised.
Speech recognitionSpeech recognition is an interdisciplinary subfield of computer science and computational linguistics that develops methodologies and technologies that enable the recognition and translation of spoken language into text by computers. It is also known as automatic speech recognition (ASR), computer speech recognition or speech to text (STT). It incorporates knowledge and research in the computer science, linguistics and computer engineering fields. The reverse process is speech synthesis.
Types of artificial neural networksThere are many types of artificial neural networks (ANN). Artificial neural networks are computational models inspired by biological neural networks, and are used to approximate functions that are generally unknown. Particularly, they are inspired by the behaviour of neurons and the electrical signals they convey between input (such as from the eyes or nerve endings in the hand), processing, and output from the brain (such as reacting to light, touch, or heat). The way neurons semantically communicate is an area of ongoing research.
LanguageLanguage is a structured system of communication that consists of grammar and vocabulary. It is the primary means by which humans convey meaning, both in spoken and written forms, and may also be conveyed through sign languages. The vast majority of human languages have developed writing systems that allow for the recording and preservation of the sounds or signs of language. Human language is characterized by its cultural and historical diversity, with significant variations observed between cultures and across time.
Language acquisitionLanguage acquisition is the process by which humans acquire the capacity to perceive and comprehend language (in other words, gain the ability to be aware of language and to understand it), as well as to produce and use words and sentences to communicate. Language acquisition involves structures, rules, and representation. The capacity to use language successfully requires one to acquire a range of tools including phonology, morphology, syntax, semantics, and an extensive vocabulary.
Convolutional neural networkConvolutional neural network (CNN) is a regularized type of feed-forward neural network that learns feature engineering by itself via filters (or kernel) optimization. Vanishing gradients and exploding gradients, seen during backpropagation in earlier neural networks, are prevented by using regularized weights over fewer connections. For example, for each neuron in the fully-connected layer 10,000 weights would be required for processing an image sized 100 × 100 pixels.
Feedforward neural networkA feedforward neural network (FNN) is one of the two broad types of artificial neural network, characterized by direction of the flow of information between its layers. Its flow is uni-directional, meaning that the information in the model flows in only one direction—forward—from the input nodes, through the hidden nodes (if any) and to the output nodes, without any cycles or loops, in contrast to recurrent neural networks, which have a bi-directional flow.
Artificial neural networkArtificial neural networks (ANNs, also shortened to neural networks (NNs) or neural nets) are a branch of machine learning models that are built using principles of neuronal organization discovered by connectionism in the biological neural networks constituting animal brains. An ANN is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain. Each connection, like the synapses in a biological brain, can transmit a signal to other neurons.
Domain nameIn the Internet, a domain name is a string that identifies a realm of administrative autonomy, authority or control. Domain names are often used to identify services provided through the Internet, such as websites, email services and more. As of 2017, 330.6 million domain names had been registered. Domain names are used in various networking contexts and for application-specific naming and addressing purposes. In general, a domain name identifies a network domain or an Internet Protocol (IP) resource, such as a personal computer used to access the Internet, or a server computer.
Language deathIn linguistics, language death occurs when a language loses its last native speaker. By extension, language extinction is when the language is no longer known, including by second-language speakers, when it becomes known as an extinct language. A related term is linguicide, the death of a language from natural or political causes, and, rarely, glottophagy, the absorption or replacement of a minor language by a major language.
Recurrent neural networkA recurrent neural network (RNN) is one of the two broad types of artificial neural network, characterized by direction of the flow of information between its layers. In contrast to uni-directional feedforward neural network, it is a bi-directional artificial neural network, meaning that it allows the output from some nodes to affect subsequent input to the same nodes. Their ability to use internal state (memory) to process arbitrary sequences of inputs makes them applicable to tasks such as unsegmented, connected handwriting recognition or speech recognition.
Acoustic modelAn acoustic model is used in automatic speech recognition to represent the relationship between an audio signal and the phonemes or other linguistic units that make up speech. The model is learned from a set of audio recordings and their corresponding transcripts. It is created by taking audio recordings of speech, and their text transcriptions, and using software to create statistical representations of the sounds that make up each word. Modern speech recognition systems use both an acoustic model and a language model to represent the statistical properties of speech.
Extinct languageAn extinct language is a language that no longer has any speakers, especially if the language has no living descendants. In contrast, a dead language is one that is no longer the native language of any community, even if it is still in use, like Latin. A dormant language is a dead language that still serves as a symbol of ethnic identity to a particular group. These languages are often undergoing a process of revitalisation. Languages that currently have living native speakers are sometimes called modern languages to contrast them with dead languages, especially in educational contexts.
Constructed languageA constructed language (shortened to a conlang) is a language whose phonology, grammar, and vocabulary, instead of having developed naturally, are consciously devised for some purpose, which may include being devised for a work of fiction. A constructed language may also be referred to as an artificial, planned or invented language, or (in some cases) a fictional language. Planned languages (or engineered languages/engelangs) are languages that have been purposefully designed; they are the result of deliberate, controlling intervention and are thus of a form of language planning.
Top-level domainA top-level domain (TLD) is one of the domains at the highest level in the hierarchical Domain Name System of the Internet after the root domain. The top-level domain names are installed in the root zone of the name space. For all domains in lower levels, it is the last part of the domain name, that is, the last non empty label of a fully qualified domain name. For example, in the domain name www.example.com, the top-level domain is .com. Speech perceptionSpeech perception is the process by which the sounds of language are heard, interpreted, and understood. The study of speech perception is closely linked to the fields of phonology and phonetics in linguistics and cognitive psychology and perception in psychology. Research in speech perception seeks to understand how human listeners recognize speech sounds and use this information to understand spoken language.
Speaker recognitionSpeaker recognition is the identification of a person from characteristics of voices. It is used to answer the question "Who is speaking?" The term voice recognition can refer to speaker recognition or speech recognition. Speaker verification (also called speaker authentication) contrasts with identification, and speaker recognition differs from speaker diarisation (recognizing when the same speaker is speaking).
Second languageA second language (L2) is a language spoken in addition to one's first language (L1). A second language may be a neighbouring language, another language of the speaker's home country, or a foreign language. A speaker's dominant language, which is the language a speaker uses most or is most comfortable with, is not necessarily the speaker's first language. For example, the Canadian census defines first language for its purposes as "the first language learned in childhood and still spoken", recognizing that for some, the earliest language may be lost, a process known as language attrition.
Endangered languageAn endangered language or moribund language is a language that is at risk of disappearing as its speakers die out or shift to speaking other languages. Language loss occurs when the language has no more native speakers and becomes a "dead language". If no one can speak the language at all, it becomes an "extinct language". A dead language may still be studied through recordings or writings, but it is still dead or extinct unless there are fluent speakers.
Time delay neural networkTime delay neural network (TDNN) is a multilayer artificial neural network architecture whose purpose is to 1) classify patterns with shift-invariance, and 2) model context at each layer of the network. Shift-invariant classification means that the classifier does not require explicit segmentation prior to classification. For the classification of a temporal pattern (such as speech), the TDNN thus avoids having to determine the beginning and end points of sounds before classifying them.
