EVOPEDIA

← all systems · see it in The System

Speech (human spoken language)

Speech is the use of the human voice as a medium for language.

Properties published scienceour proposal

Sourced fields, and our own readings.

born atcontested: anatomy for contrasting vowels over ~20 Ma (Boë 2019); full human speech with modern humans; the laryngeal descent theory's ~200 ka
born precisionunresolved
boundaryInside: what a person does to speak and to take in speech: the breath, larynx and vocal tract that make the voice and shape it into vowels and consonants; the nerves and brainstem nuclei that drive over 100 muscles; the brain's speech networks that plan, monitor and correct speech and turn heard sound into speech sounds and words; and the timing of talk between people. Outside: the language that speech carries (its grammar and lexicon: the language walk), the sound medium itself (acoustic-wave), the hearing apparatus as a sense (human's hearing channel, ear), sign and writing as other modalities, and the phonetic description of the sounds (phonetics, sibling).
nested levelA capacity and activity of a person (part of the human, on Human's speech outlet and hearing channel), run by parts of the human brain and the upper airway; language, conversation and speech communities run on it.
scaleorganism

Measures (24) published science

information rate39.15 (SD 5.10) bits/s
syllable rate (characteristic clock)6.63 (SD 1.15) syllables/s
speech modulations tracked by theta neurons5-10 Hz
auditory integration windows~150 ms and above; ~30 ms and below
typical gap between turns100-300 ms
speaker change in adult conversation~250 ms
preschool response latency0.8-1.8 s
word production latency (picture naming)600 (conceptual 175, lemma 75, phonological code 80, syllabification 125, phonetic encoding 145) ms
sentence planning before speech begins~1500 ms
natural conversation speed160 words per minute
speech BCI decoding speed62 words per minute
speech BCI word error rate9.1% (50 words); 23.8% (125,000 words)
muscles coordinated in speech>100 muscles
perceptual narrowing window6-12 months of age
native vowel contrasts learned~6 months of age
babbling onset4-6 months of age
first word~1 year of age
speech sounds mastered~8 years of age
syllables heard in the Mandarin exposure study~33,000 (4 talkers, 4-5 weeks) syllables
shift in the dawn of speechfrom ~200 ka to over ~20 Ma years ago
Kebara 2 Neanderthal age~60 ka
birds-humans divergence (convergent vocal learning)>310 million years
FOXP2 human-chimpanzee difference2 amino acids
gap marking a dispreferred reply>=700 ms

Open contentions (11) open contention

Where sources or readings disagree. Both stay on the map.

Cross-scale patterns seen here (15) our proposal

Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Speech (human spoken language). Each links to every other system that shows it.

Valence: what it trades, and with whom (15) published science

Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.

Partners and parts published science

Built from / builds (13)

Sits inside / holds (4)

expressed by (18)

floor filled by (28)

Kinds (8)

Notes published science

Layers

The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.

genes for speech: FOXP2 and the vocal-learning circuit

  1. L1: FOXP2 on chromosome 7q31 (the SPCH1 interval): a point mutation in the KE family and a translocation in CS disrupt speech and language
  2. L2: A forkhead-box transcription factor; two functional copies seem to be needed to acquire normal spoken language
  3. L3: Human FOXP2 differs from the chimpanzee's by two amino acids (T303N, N325S); a signal of recent selection is reported (contested)
  4. L4: Convergent gene-expression specializations in the brain regions for learned vocalization, shared by humans and song-learning birds

speech learned in childhood: from heard speech to inner speech

  1. L1: Speech around the child, from before birth: the mother's voice preferred by newborns; tens of thousands of syllables heard from live talkers
  2. L2: Perception narrows from all phonetic contrasts to the native ones between 6 and 12 months: vowels by about 6 months, consonants by 11-12 months
  3. L3: Canonical babbling: repeated consonant-vowel syllables (/bababa/, /badada/), from about 4-6 months; by 10 months babbling differs by the language heard
  4. L4: First words at about 1 year, multiword utterances in the second year
  5. L5: The native sound system mastered by about age 8; turn timing keeps developing through childhood
  6. L6: Social speech turns inward: private speech in young children, then inner speech (Vygotsky's internalization)

speech: the speech chain from speaker to listener (reader's view)

  1. L1: Speech sound in the air: an alternation of sound and silence, carrying voice and formant changes from the speaker's mouth to every listener in range and back to the speaker's own ears
  2. L2: The vocal apparatus and the ear: expiratory airflow (power), the vocal folds (source), the tongue, jaw, lips and pharynx (filter); on the receiving side the auditory periphery
  3. L3: Nerves and brainstem: the vagus nerve's superior and recurrent laryngeal branches, motoneurons in the nucleus ambiguus (larynx) and the trigeminal motor, facial and hypoglossal nuclei (articulators), and the expiratory motoneurons, reached through the reticular formation
  4. L4: Sensory and motor cortex for speech: dorsal superior temporal gyrus and sulcus (spectrotemporal analysis and phonological representation, bilateral); ventral sensorimotor cortex with the laryngeal motor cortex
  5. L5: The speech networks: a ventral stream for comprehension and a strongly left-dominant dorsal stream through area Spt to the frontal lobe for sensory-motor integration; feedback and feedforward control (premotor cortex, cerebellum), the SMA and basal ganglia gating when commands are released
  6. L6: Words and plans: lexical access, phonological code retrieval, syllabification and phonetic encoding (about 600 ms to name a picture); inner speech, the same machinery without sound
  7. L7: Talk between people: turns at talking exchanged with gaps of a few hundred milliseconds, speech acts (informing, asking, persuading), and the timing that makes a reply readable as willing or reluctant

the boardroom (shared floors)

  1. L5: The speech networks: a largely bilateral ventral stream for comprehension, a strongly left-dominant dorsal stream (Spt to frontal cortex) for sensory-motor integration; DIVA's feedback control map in right ventral premotor cortex; the SMA and basal ganglia gating release of commands, the cerebellum tuning feedforward commands; separate basal ganglia populations for speaking and for ambient noise
  2. L6: Words and plans: lemma retrieval, phonological code retrieval, syllabification and phonetic encoding into the speech sound map (left ventral premotor and inferior frontal cortex), which holds a motor program for each frequent sound
  3. L6: Inner speech: the internal monologue or dialogue, expanded or condensed, used to regulate thought and behaviour; associated with left inferior frontal gyrus, left insula and left STG
  4. L7: Talk between people: turn-taking with the next turn projected from the incoming one, replies planned before the other has finished, overlaps resolved by one speaker dropping out

voice: breath, larynx and vocal tract (source and filter)

  1. L1: Expiratory airflow from the lungs, the power source; speech breathing uses shorter, deeper inspirations and long exhalations, taken at phrase boundaries
  2. L2: The vocal folds in the glottis: brought close so that airflow sets them vibrating (voicing); their tension sets the pitch; fluid-structure interaction controlled by laryngeal muscles
  3. L3: The vocal tract as a filter: vowels need control of the lip opening and of where and how tightly the tongue narrows the tract; the resonances (formants) carry most phonetic information
  4. L4: Radiated speech: a stable, harmonic-rich source with rapidly changing formants

way in: Hearing: ambient noise

  1. L1: Background noise in the room or street, competing with speech for the listener's ear

way in: Hearing: one's own voice (auditory feedback)

  1. L1: The speaker's own speech coming back through the air
  2. L4: Auditory target and error maps on the posterior superior temporal gyrus (near the STS and at the temporo-parietal junction, area Spt), where the expected sound inhibits the matching part of the error map

way in: Hearing: others' speech (the most typical channel)

  1. L1: Connected speech in the air, organised at several time scales at once (fast fine structure, syllables, phrases)
  2. L3: Thalamo-cortical input arriving in layer IV of auditory cortex
  3. L4: Auditory cortex samples speech in two windows, ~150 ms (syllables, theta) and ~30 ms (phonemic detail, gamma), discarding the beta range

way in: Inner: emotion and arousal (the limbic vocal route)

  1. L1: Contested origin (C-413, as Human's emotion channel). In the vocal system, emotional state drives innate vocalizations through the limbic vocal control pathway
  2. L4: Anterior cingulate cortex and the periaqueductal grey, converging with the laryngeal motor cortex pathway in the ACC and the brainstem reticular formation

way in: Inner: what there is to say (the message)

  1. L1: Contested: where the thought to be said comes from is not settled (C-413). In the models read, input to speech comes from a lexical-conceptual network distributed through cortex
  2. L4: Conceptual preparation: from an intention to the concept to be named (about 175 ms of the 600 ms in picture naming)

way in: Sight: the talker's face and lips

  1. L1: Light from the talker's moving lips and face; visual information enters the decision about which sound was heard (the McGurk effect)

way in: Touch and stretch of one's own lips, tongue, jaw and vocal folds (somatosensory feedback)

  1. L1: Mechanical displacement of perioral tissue, contact of tongue and palate, stretch of the vocal folds as they move
  2. L2: Orofacial and laryngeal mechanoreceptors (vocal fold somatosensation, tongue-tip touch)
  3. L4: Somatosensory target and state maps in ventral supramarginal gyrus: the expected proprioceptive and tactile feedback for each sound

way out: Innate vocalizations: cry, laughter, emotional calls (the limbic route)

  1. L1: Cry, shriek and laughter in the air: genetically preprogrammed acoustic structure
  2. L4: Brainstem vocal patterning, enough on its own for vocal reactions to pain
  3. L5: The limbic vocal control pathway, from the anterior cingulate cortex down through the periaqueductal grey

way out: Inner speech (speaking without sound)

  1. L1: Verbal mediation of short-term memory, used by children from about 7 years
  2. L6: Imagined speech with the timing and slips of real speech
  3. L7: Self-regulation of thought and behaviour

way out: Speaking aloud

  1. L1: Speech in the air, louder in noise so that listeners receive it
  2. L2: Lungs, larynx (glottis) and vocal tract: pulmonic speech
  3. L3: Cranial nerves: the vagus nerve's superior and recurrent laryngeal branches to the larynx
  4. L4: Brainstem motoneurons: laryngeal (nucleus ambiguus), articulatory (trigeminal motor, facial, hypoglossal) and expiratory, reached directly from LMC and through the reticular formation
  5. L5: Primary motor cortex: articulator maps in ventral precentral gyrus and the laryngeal motor cortex in area 4 (dorsal LMC encoding pitch for intonation)
  6. L6: Phonetic encoding: a phonological input turned into feedforward motor commands in the speech sound map
  7. L7: The next turn, begun before the incoming one ends

way out: Speech decoded by a brain-computer interface

  1. L1: Text on a screen at 62 words per minute (9.1% word errors on 50 words; 23.8% on 125,000)
  2. L5: Attempted speech read from area 6v (ventral premotor cortex), whose neurons are tuned to articulators, phonemes and words even years after paralysis

Who fills each layer

Sources (36) published science

Speech (Wikipedia, rev 1381777470, 2026-10-24)
Tourville JA, Guenther FH. 2011. The DIVA model: a neural theory of speech acquisition and production. doi:10.1080/01690960903498424 (behind its links)
Bullock L et al. 2026. Independent basal ganglia neural populations encode speech production and ambient-noise levels. doi:10.1038/s41467-026-77937-8 (behind its links)
Kuhl PK (2010) Brain mechanisms in early language acquisition. Neuron 67:713-727. doi:10.1016/j.neuron.2010.08.038 (author manuscript) (behind its links)
Levinson SC, Torreira F. 2015. Timing in turn-taking and its implications for processing models of language. doi:10.3389/fpsyg.2015.00731 (behind its links)
Giraud A-L, Poeppel D. 2012. Cortical oscillations and speech processing: emerging computational principles and operations. doi:10.1038/nn.3063 (behind its links)
Alderson-Day B, Fernyhough C. 2015. Inner speech: development, cognitive functions, phenomenology, and neurobiology. doi:10.1037/bul0000021 (behind its links)
Fitch WT, de Boer B, Mathur N, Ghazanfar AA (2016) Monkey vocal tracts are speech-ready. Science Advances 2:e1600723. doi:10.1126/sciadv.1600723 (behind its links)
Hickok G, Houde J, Rong F. 2011. Sensorimotor integration in speech processing: computational basis and neural organization. doi:10.1016/j.neuron.2011.01.019 (behind its links)
Phonetics (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Intonation (linguistics) (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Anikin A, Reby D, Pisanski K. 2025. Nonlinear vocal phenomena and speech intelligibility. doi:10.1098/rstb.2024.0254 (behind its links)
Willett FR et al. 2023. A high-performance speech neuroprosthesis. doi:10.1038/s41586-023-06377-x (behind its links)
Coupé C, Oh YM, Dediu D, Pellegrino F (2019) Different languages, similar encoding efficiency: Comparable information rates across the human communicative niche. Science Advances 5: eaaw2594. doi:10.1126/sciadv.aaw2594 (CC BY-NC; PMC6984970) (behind its links)
Takahasi M, Mazuka R. 2026. Evolutionary background of precise vocal-respiratory temporal coordination in human infants. doi:10.1111/nyas.70409 (behind its links)
Simonyan K, Horwitz B. 2011. Laryngeal motor cortex and control of speech in humans. doi:10.1177/1073858410386727 (behind its links)
Speech perception (Wikipedia, rev 1374682269, 2026-09-13) (behind its links)
Boë LJ, Sawallis TR, Fagot J, et al. (2019) Which way to the dawn of speech?: Reanalyzing half a century of debates and data in light of speech science. Science Advances 5:eaaw3916. doi:10.1126/sciadv.aaw3916 (behind its links)
D'Anastasio R et al. 2013. Micro-biomechanics of the Kebara 2 hyoid and its implications for speech in Neanderthals. doi:10.1371/journal.pone.0082261 (behind its links)
Pfenning AR et al. 2014. Convergent transcriptional specializations in the brains of humans and song-learning birds. Science 346:1256846. doi:10.1126/science.1256846 (behind its links)
FOXP2 (Wikipedia, rev 1379237867, 2026-10-08) (behind its links)
Dichter BK, Breshears JD, Leonard MK, Chang EF. 2018. The control of vocal pitch in human laryngeal motor cortex. doi:10.1016/j.cell.2018.05.016 (behind its links)
Hickok G, Poeppel D. 2007. The cortical organization of speech processing. doi:10.1038/nrn2113 (behind its links)
Public speaking (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Stephens, Silbert, Hasson 2010. Speaker-listener neural coupling underlies successful communication. PNAS 107:14425. doi:10.1073/pnas.1008662107 (PMC2922522) (behind its links)
DeCasper AJ, Fifer WP. 1980. Of human bonding: newborns prefer their mothers' voices. doi:10.1126/science.7375928 (behind its links)
Fuchs S et al. 2026. Toward a fuller integration of respiratory rhythms into research on infant vocal and motor development. doi:10.1111/nyas.70197 (behind its links)
Indefrey P. 2011. The spatial and temporal signatures of word production components: a critical update. Front Psychol 2:255. doi:10.3389/fpsyg.2011.00255 (behind its links)
Linguistics (encyclopedia article, WME snapshot 2026-11-01) (behind its links)
Stivers T et al. (2009) Universals and cultural variation in turn-taking in conversation. PNAS 106(26): 10587-10592 (abstract read via ICCI blog) (behind its links)
Nishimura T et al. 2022. Evolutionary loss of complexity in human vocal anatomy as an adaptation for speech. doi:10.1126/science.abm1574 (behind its links)
Lai CS et al. 2001. A forkhead-domain gene is mutated in a severe speech and language disorder. doi:10.1038/35097076 (behind its links)
Enard W et al. 2002. Molecular evolution of FOXP2, a gene involved in speech and language. doi:10.1038/nature01025 (behind its links)
Wang H, Max L. 2026. Disorders of the basal ganglia and cerebellum: a comprehensive review of speech sensorimotor symptoms and responses. doi:10.1162/nol.a.291 (behind its links)
Fitch WT, Reby D. 2001. The descended larynx is not uniquely human. doi:10.1098/rspb.2001.1704 (behind its links)
Smith A, Weber C. 2017. How stuttering develops: the multifactorial dynamic pathways theory. doi:10.1044/2017_JSLHR-S-16-0343 (behind its links)