flite

List all available voices

$ flite -lv

$ flite -t "[string]"

$ flite -f [path/to/file.txt]

$ flite -voice [file://path/to/filename.flitevox|url]

$ flite -voice [file://path/to/filename.flitevox|url] -f [path/to/file.txt] -o [output.wav]

$ flite --version

flite [options] [files]

-t
    Treat all input as plain text (default: SSML)

-f <file>
    Read input from specified file

-o <file>
    Write WAV output to file (default: stdout)

-voice <name>
    Select voice (e.g., kal, rms, awb)

--help
    Display usage summary

--version
    Show Flite version

-ssml
    Explicitly set SSML input mode (default)

Flite (Festival Lite) is a compact, fast runtime speech synthesis engine developed by Carnegie Mellon University. Designed for embedded systems and resource-constrained environments, it converts text or SSML input into speech audio. Flite supports multiple voices like kal, rms, and others, outputting to WAV files or stdout for playback. It processes plain text by default in SSML mode but can handle direct text with options. Ideal for scripting, alerts, or accessibility tools, Flite requires minimal dependencies compared to full Festival. Voices are selected via -voice and often installed separately. It streams output efficiently, supporting real-time synthesis on Linux systems.

Requires flite voices package for full voice support; limited to installed voices. No GUI; audio playback needs external player like aplay(1). CPU-intensive for long texts on low-end hardware.

flite -t "Hello, world!" (speak text)
flite -t "Test" -o test.wav (save to WAV)
flite -voice kal -f script.txt (use file and voice)

Run flite -lv to list available voices. Common: kal (female), rms (male), slt.

Originated in 2001 from Festival project at CMU for lightweight TTS. Evolved for mobile/embedded use; current versions (2.x) add better voice support and SSML.

festival(1), espeak(1), pico2wave(1)