Guides · 6 min read
How LaTeX is read aloud: a reference for common commands
LaTeX describes how an equation looks; reading it aloud means saying what it means. \frac{a}{b} becomes “a over b”, x_i becomes “x sub i”, and \sum_{i=1}^{n} becomes “the sum from i equals 1 to n of”. The tables list the spoken form of common commands, produced by the engine that narrates books in Aethel Audio.
Updated 11 November 2026
Fractions, powers and roots
A fraction is “over”. A longer numerator or denominator is fenced with “the quantity … all over”, so a listener hears where it starts and ends.
| LaTeX | Read aloud as |
|---|---|
| \frac{a}{b} | a over b |
| \frac{x + 1}{x - 1} | the quantity x plus 1, all over the quantity x minus 1 |
| x^2 | x squared |
| x^3 | x cubed |
| x^{n+1} | x to the power of n plus 1 |
| e^{-x} | e to the power negative x |
| \sqrt{x} | the square root of x |
| \sqrt[3]{x} | the cube root of x |
| x_i | x sub i |
| a_{ij} | a sub i j |
Sums, products, integrals and limits
The limits are read before the thing being added or integrated: “from … to … of”.
| LaTeX | Read aloud as |
|---|---|
| \sum_{i=1}^{n} x_i | the sum from i equals 1 to n of x sub i |
| \prod_{i=1}^{n} x_i | the product from i equals 1 to n of x sub i |
| \int_a^b f(x)\,dx | the integral from a to b of f of x, d x |
| \int x^2 \, dx | the integral of x squared d x |
| \oint_C \mathbf{F} \cdot d\mathbf{r} | the contour integral over C of F dot d r |
| \lim_{x \to \infty} f(x) | the limit as x goes to infinity of f of x |
| \sum_{k=0}^{\infty} \frac{x^k}{k!} | the sum from k equals 0 to infinity of x to the power k over k factorial |
Derivatives and vector calculus
A derivative written as a fraction is not read as a fraction, and ∇ is named by what it does.
| LaTeX | Read aloud as |
|---|---|
| \frac{dy}{dx} | the derivative of y with respect to x |
| \frac{\partial f}{\partial x} | the partial derivative of f with respect to x |
| f^{\prime}(x) | f prime of x |
| \dot{x} | x dot |
| \nabla f | the gradient of f |
| \nabla \cdot \mathbf{F} | the divergence of F |
| \nabla \times \mathbf{F} | the curl of F |
Vectors and matrices
Accents are said after the letter (“x bar”, “y hat”); a transpose and an inverse are named.
| LaTeX | Read aloud as |
|---|---|
| \hat{y} | y hat |
| \bar{x} | x bar |
| \vec{v} | vector v |
| A^\top | A transpose |
| \mathbf{A}^{-1} | A inverse |
| \det(A) | the determinant of A |
| \|x\| | the norm of x |
| \|x\|_2 | the 2 norm of x |
| \begin{pmatrix} a & b \\ c & d \end{pmatrix} | the matrix with rows a, b; c, d |
Sets, logic and relations
Relations are read as short phrases, so the sentence still parses: “is less than or equal to”, “is a subset of”.
| LaTeX | Read aloud as |
|---|---|
| x \in A | x in A |
| A \subseteq B | A is a subset of B |
| A \cup B | A union B |
| A \cap B | A intersection B |
| \forall x | for all x |
| \exists x | there exists x |
| \mathbb{R} | the real numbers |
| a \leq b | a is less than or equal to b |
| a \neq b | a is not equal to b |
| a \approx b | a is approximately b |
| a \propto b | a is proportional to b |
| f \circ g | f composed with g |
Probability, statistics and functions
A bar inside a probability is “given”; a tilde before a distribution is “is distributed as”.
| LaTeX | Read aloud as |
|---|---|
| P(A \mid B) | P of A given B |
| \mathbb{E}[X] | the expected value of X |
| \operatorname{Var}(X) | the variance of X |
| x \sim \mathcal{N}(\mu, \sigma^2) | x is distributed as N of mu, sigma squared |
| \binom{n}{k} | n choose k |
| n! | n factorial |
| \log_2 n | log base 2 of n |
| \ln x | natural log x |
| \sin \theta | sine theta |
| \arg\max_x f(x) | the arg max over x of f of x |
| g^{-1}(y) | g inverse of y |
The rules behind the tables
- Say structure in words. A listener cannot see brackets or a fraction bar, so grouping is spoken: “the quantity”, “all over”, “of”.
- Name symbols by meaning, not shape: \nabla \cdot \mathbf{F} is “the divergence of F”, not “nabla dot F”.
- Keep one phrase for one construction. A subscript is always “sub”, a power is “squared”, “cubed” or “to the power”.
- Read relations as verbs so the sentence parses: “x is less than or equal to y”.
- Leave formatting silent. Bold, upright and calligraphic letters are read as the letter, because the context says what they are.
Where context decides
- Bars: |x| is “the absolute value of x”, but P(A | B) is “P of A given B”.
- A raised −1: on a function it is the inverse function, “f inverse of x”; on a matrix it is “A inverse”; on a number it is a power.
- Primes: f′ is “f prime”, which is usually a derivative but can be just another function.
- A tilde: x ∼ N(μ, σ²) is “is distributed as”; elsewhere ∼ is “is similar to”.
- A capital letter raised above another, as in W^Q, is a label, not a power.
An engine has to make these calls from the surrounding notation, and it can get them wrong. That is why Aethel keeps the printed equation on screen while it reads.
Convert your own LaTeX
Paste any equation into the free LaTeX to speech converter to get its spoken form. It is useful for writing alt text, for lecture notes that will be read aloud, and for checking how a formula will sound before you record.
Questions
Is there one correct way to read an equation aloud?
No. Lecturers differ, and so do conventions between countries and fields. What matters is that the reading is unambiguous: a listener should be able to write the equation down from what they hear.
How do you say \frac{a}{b} in LaTeX?
“a over b”. With longer parts, the fraction is fenced so it is clear where it ends: \frac{x + 1}{x - 1} is “the quantity x plus 1, all over the quantity x minus 1”.
Can a screen reader read LaTeX?
A screen reader reads raw LaTeX character by character, backslashes included. Pages that render LaTeX with MathJax or as MathML expose the structure instead, and are read as maths.
Does Aethel read LaTeX files?
Aethel reads books and papers as PDF, EPUB or text, and rebuilds the equations in them as LaTeX before narrating. For a single LaTeX equation, use the converter.