say

Speak text aloud

$ say "[Hello, world]"

$ say -v [Alex] "[Hello]"

$ say -v "?"

$ say -f [input.txt]

$ say -o [output.aiff] "[Hello, world]"

$ say -r [200] "[Speaking faster now]"

$ say -a "[Built-in Output]" "[Hello]"

say [-v voice] [-r rate] [-o outfile] [-f infile] [-a device] [text...]

say converts text to audible speech using the macOS Speech Synthesis manager. It speaks text provided as arguments, from a file, or from standard input, either through audio output or saved to a file.
Multiple voices are available with different accents, languages, and personalities. The default voice and speech settings can be configured in System Settings under Accessibility > Spoken Content.
Output can be saved to AIFF, WAVE, m4af, or caff audio formats for later playback or processing.

-v voice

Use the specified voice instead of the system default

Speech rate in words per minute

Read text from the specified file

Save speech to an audio file (AIFF, WAVE, m4af, or caff format)

Specify audio output device; use '?' to list devices

Send speech to a remote audio server

Interactive mode for continuous text input

Direct MP3 output is not supported; convert AIFF output using a separate tool. Voice availability depends on what is installed in macOS; additional voices can be downloaded from System Settings.

say has been included in macOS since the early days of Mac OS X, building on Apple's text-to-speech technology that dates back to the 1980s with MacinTalk. The command provides terminal access to the same speech synthesis used throughout the operating system.

espeak(1), festival(1), afplay(1)