Free reference
Viseme 치트 시트
A visual guide for artists and researchers working on lipsync: American English phonemes and the visemes they commonly map to.
Phoneme to viseme chart
English consonants. Hover over a sound to see its lip shape. Select it to jump to its photo poses.
In a pair, the voiceless sound is on the left. To hear each phoneme, visit the IPA chart hosted by Cambridge University Press & Assessment.
Chart Key Terms
bilabial: the upper and lower lips come together
labiodental: the lower lip touches or comes close to the upper teeth (the reverse, upper lip to lower teeth, is possible but rare)
dental: the tongue tip touches or comes close to the teeth, either between the upper and lower teeth or behind the upper teeth
alveolar: the tongue tip touches or comes close to the gum ridge just behind the upper teeth (the alveolar ridge)
postalveolar: the tongue makes a narrow passage just behind the alveolar ridge
palatal: the middle of the tongue rises toward the hard palate, the bony center of the roof of the mouth
velar: the back of the tongue touches or comes close to the 벨룸(연한 입천장)
glottal: the sound is made at the 성문, the opening between the vocal folds: closing it briefly makes ʔ. For h, the vocal folds are held apart to allow free airflow
labiovelar: two constrictions at once: the lips round while the back of the tongue rises toward the velum, as in w. Neither closes completely, which makes it an approximant (more on approximants)
Before you copy a pose
These are isolated poses. Visemes in natural speech are much more complicated and context-dependent, but being able to observe them in unobstructed form is a helpful first step.
Every viseme varies. Some have fixed requirements (e.g. m/b/p need the lips to close entirely or nearly entirely), but the facial actions fluctuate with individual facial features, coarticulation, emotional context, volume, and other speech-related conditions. More on variability
Consonant poses
Ordered from the lips back to the throat, like the chart. Select a sound to switch the photo.
중립/무음
소리 없음, 편안함
p/b/m
p 에서와 같이 pout
bilabial, plosive, voiceless
/p/ and /b/ require the lips to part after the initial closure to propel air for the plosive sound, hence the closed and open states.
b 에서와 같이 buccinator
양순, 파열음, 유성음
/p/ and /b/ require the lips to part after the initial closure to propel air for the plosive sound, hence the closed and open states.
m 에서와 같이 m자산
양순, 비강, 유성음
f/v
f 에서와 같이 f에이스
치조음, 마찰음, 무성음
v 에서와 같이 v시각적
치조음, 마찰음, 유성음
θ/ð (th)
θ 에서와 같이 thoracic, wreath
치아, 마찰음, 무성음
ð as in teething, mouthing
치과, 마찰음, 유성음
t/d
t 에서와 같이 t엠포랄리스
폐포, 파열음, 무성음
d 에서와 같이 d압착기
폐포, 파열음, 유성음
Under most circumstances, ɾ as in butter may be included in this viseme group.
s/z
s 에서와 같이 see
폐포, 마찰음, 무성음
z 에서와 같이 zygomatic
폐포, 마찰음, 유성음
n/l
n 에서와 같이 nasal
폐포, 비강, 유성음
l 에서와 같이 labial
폐포, 측면, 유성음
tʃ/dʒ/ʃ/ʒ
tʃ 에서와 같이 chin
postalveolar, affricate, voiceless
dʒ 에서와 같이 jaw, gem
postalveolar, affricate, voiced
ʃ 에서와 같이 shine, 스파tial
postalveolar, fricative, voiceless
ʒ 에서와 같이 g엔레, 메아sure
postalveolar, fricative, voiced
w
w 에서와 같이 wink
labiovelar, approximant, voiced
Under most circumstances, ɹ at the start of a syllable, as in red, may be included in this viseme group.
k/g
k 에서와 같이 king
모음, 파열음, 무성음
g 에서와 같이 guess
모음, 파열음, 유성음
Under most circumstances, ŋ as in tongue and ʔ as in uh-oh may be included in this viseme group.
In many circumstances, j as in yawn may also be included in this viseme group.
h
h 에서와 같이 h아이포캠퍼스, happy
성문, 마찰음, 무성음
Vowel poses
i (ee)
에서와 같이 eel, cheese, meat
ɪ (ih)
에서와 같이ip, which
ɛ (eh)
에서와 같이 aNY, Kept, levator
æ
에서와 같이at, FACS, zygoma티쿠스
ɑ (아)
에서와 같이aw, cauSE, STop
This reference doesn’t distinguish between [ɑ] and [ɔ].
ə (어)
에서와 같이 anatomy, nasal, about
ʊ (ouh)
에서와 같이uller, look, good
u (oo)
에서와 같이ood, rue, mood
oʊ (oh)
에서와 같이oWERER, SLow, moat
ɔɪ (oy)
에서와 같이oy, soy, join
The basics
비젬이란 무엇인가요?
A viseme is a group of phonemes that are visually indistinct from one another. /f/ and /v/ are acoustically distinct, but their observable lip, tongue, and/or jaw configurations are not perceptibly different, so they map to the same viseme.
The word is a portmanteau of “visual” and “phoneme,” coined by 클레투스 G. 피셔 그의 중요한 1968년 논문에서 "시각적으로 인식되는 자음의 혼동.”
"시각적 음소라는 문구는 시각적 음소로 줄였으며, 개별적이고 대조적인 시각적 인식 단위를 지칭하는 데 사용될 것입니다."
음소는 어떻게 비셈에 매핑되나요?
It’s a murky area in facial animation and research. There is no standardized set of visemes and no standardized method for grouping phonemes, so you will find different mappings everywhere you go.
How you should cluster phonemes depends heavily on your use case. If you want simple speech movements or can’t represent the tongue, you do not have a strong need to differentiate things like /t/ vs. /k/. If you want to get as nuanced as possible and can represent key tongue positions, it would behoove you to consider a more complex grouping system.
How is “viseme” pronounced?
Fisher did not specify, so there is no formal record. You may hear “veye-zeem,” “vih-zeem,” “veye-seem,” “vih-seem,” and others. Personally, I prefer “veye-zeem,” but I’m open to considering arguments against it.
For structured learning and frameworks
Live workshop · March 6, 2027
Lipsync Mechanics
Reading phonemes, building visemes, and judging lipsync. Learn the mechanical requirements for each sound as well as why and how to build them with FACS-based shapes.
Go deeper on speech
More reference guides
How to cite this page
APA
오젤, M. (2022, 1월). Viseme 치트 시트. FACS를 마주합니다. https://melindaozel.com/viseme-cheat-sheet/
BibTeX
@misc{ozel2022viseme,
저자 = {오젤, 멜린다},
title = {Viseme 치트 시트},
year = {2022},
month = jan,
howpublished = {\url{https://melindaozel.com/viseme-cheat-sheet/}}
}