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.
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.
Phonemic orthographyA phonemic orthography is an orthography (system for writing a language) in which the graphemes (written symbols) correspond to the phonemes (significant spoken sounds) of the language. Natural languages rarely have perfectly phonemic orthographies; a high degree of grapheme–phoneme correspondence can be expected in orthographies based on alphabetic writing systems, but they differ in how complete this correspondence is.
Multilayer perceptronA multilayer perceptron (MLP) is a misnomer for a modern feedforward artificial neural network, consisting of fully connected neurons with a nonlinear kind of activation function, organized in at least three layers, notable for being able to distinguish data that is not linearly separable. It is a misnomer because the original perceptron used a Heaviside step function, instead of a nonlinear kind of activation function (used by modern networks).
Speech sound disorderA speech sound disorder (SSD) is a speech disorder in which some sounds (phonemes) are not produced or used correctly. The term "protracted phonological development" is sometimes preferred when describing children's speech, to emphasize the continuing development while acknowledging the delay. Speech sound disorders may be subdivided into two primary types, articulation disorders (also called phonetic disorders) and phonemic disorders (also called phonological disorders).
Articulatory phoneticsThe field of articulatory phonetics is a subfield of phonetics that studies articulation and ways that humans produce speech. Articulatory phoneticians explain how humans produce speech sounds via the interaction of different physiological structures. Generally, articulatory phonetics is concerned with the transformation of aerodynamic energy into acoustic energy. Aerodynamic energy refers to the airflow through the vocal tract. Its potential form is air pressure; its kinetic form is the actual dynamic airflow.
Voice (phonetics)Voice or voicing is a term used in phonetics and phonology to characterize speech sounds (usually consonants). Speech sounds can be described as either voiceless (otherwise known as unvoiced) or voiced. The term, however, is used to refer to two separate concepts: Voicing can refer to the articulatory process in which the vocal folds vibrate, its primary use in phonetics to describe phones, which are particular speech sounds. It can also refer to a classification of speech sounds that tend to be associated with vocal cord vibration but may not actually be voiced at the articulatory level.
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.
Feature learningIn machine learning, feature learning or representation learning is a set of techniques that allows a system to automatically discover the representations needed for feature detection or classification from raw data. This replaces manual feature engineering and allows a machine to both learn the features and use them to perform a specific task. Feature learning is motivated by the fact that machine learning tasks such as classification often require input that is mathematically and computationally convenient to process.
English phonologyEnglish phonology is the system of speech sounds used in spoken English. Like many other languages, English has wide variation in pronunciation, both historically and from dialect to dialect. In general, however, the regional dialects of English share a largely similar (but not identical) phonological system. Among other things, most dialects have vowel reduction in unstressed syllables and a complex set of phonological features that distinguish fortis and lenis consonants (stops, affricates, and fricatives).
Language modelA language model is a probabilistic model of a natural language that can generate probabilities of a series of words, based on text corpora in one or multiple languages it was trained on. Large language models, as their most advanced form, are a combination of feedforward neural networks and transformers. They have superseded recurrent neural network-based models, which had previously superseded the pure statistical models, such as word n-gram language model.
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.
Perceptrons (book)Perceptrons: an introduction to computational geometry is a book written by Marvin Minsky and Seymour Papert and published in 1969. An edition with handwritten corrections and additions was released in the early 1970s. An expanded edition was further published in 1987, containing a chapter dedicated to counter the criticisms made of it in the 1980s. The main subject of the book is the perceptron, a type of artificial neural network developed in the late 1950s and early 1960s.
AllophoneIn phonology, an allophone (ˈæləfoʊn; from the Greek ἄλλος, , 'other' and φωνή, , 'voice, sound') is one of multiple possible spoken sounds - or phones - or signs used to pronounce a single phoneme in a particular language. For example, in English, the voiceless plosive t (as in stop [ˈstɒp]) and the aspirated form th (as in top [ˈthɒp]) are allophones for the phoneme /t/, while these two are considered to be different phonemes in some languages such as Thai.
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.
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.
H-droppingH-dropping or aitch-dropping is the deletion of the voiceless glottal fricative or "H-sound", [h]. The phenomenon is common in many dialects of English, and is also found in certain other languages, either as a purely historical development or as a contemporary difference between dialects. Although common in most regions of England and in some other English-speaking countries, and linguistically speaking a neutral evolution in languages, H-dropping is often stigmatized as a sign of careless or uneducated speech.
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.
PhonemeIn phonology and linguistics, a phoneme (ˈfoʊniːm) is a unit of phone that can distinguish one word from another in a particular language. For example, in most dialects of English, with the notable exception of the West Midlands and the north-west of England, the sound patterns sɪn (sin) and sɪŋ (sing) are two separate words that are distinguished by the substitution of one phoneme, /n/, for another phoneme, /ŋ/. Two words like this that differ in meaning through the contrast of a single phoneme form a minimal pair.
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.
