Choose the amount of text you need to hear. Use a line or paragraph for a local check, a selected region for exact editing boundaries, or C-e i for longer reading with a tracked position. Ordinary movement speaks what you reach; explicit inspection does not require that movement first.
The default keys are collected here. The following two sections explain which preserve point, how to stop and start again, and the limits of large requests. These distinctions matter before you begin selecting text.
emacsvox-speak-char
speaks the character at point phonetically.
emacsvox-speak-word
speaks the current word. Repeating it at the same position spells the word.
emacsvox-speak-line
speaks the current line.
emacsvox-speak-paragraph
speaks the current paragraph.
emacsvox-speak-page
speaks the current page.
emacsvox-speak-region
speaks the region between point and mark. Establish and check the selection
before editing; large requests have the windowful limit explained below.
emacsvox-speak-rest-of-buffer
reads from point towards the buffer end with chunk-based position tracking.
It requires a supporting server; see Read from point, stop, and start again.
emacsvox-speak-buffer
reads the accessible buffer without tracking point through the speech.
Large requests fall back to windowful reading, as explained below.
tts-stop
stops current speech.
emacsvox-toggle-show-point
toggles the aural indication of point within spoken text.
Many reading commands accept a prefix to limit speech before or after point. Use C-h k before relying on a variant; exact prefix behaviour belongs to that command’s documentation.