speech
Generate speech audio from text using AI text-to-speech models.
Supports OpenAI (tts-1, tts-1-hd, gpt-4o-mini-tts), ElevenLabs
(eleven_v3, eleven_multilingual_v2, eleven_flash_v2_5), and Cartesia
(sonic-3.5, sonic-3). Uses your provider API key directly (egaki
subscription does not cover speech yet).
Usage
1egaki speech [text]
Arguments
| Argument | Required | Description |
[text] | No | text |
Options
| Option | Default | Description |
-m, --model [model] | - | Speech model ID. If omitted, shows an interactive picker (or uses default in non-TTY mode) |
-o, --output [path] | egaki-output | Output file path (extension added from audio format if omitted) |
--voice [voice] | - | Voice ID or name (provider-specific). OpenAI: alloy, echo, nova, etc. ElevenLabs/Cartesia: voice ID from their library |
--output-format [format] | - | Audio output format: mp3, wav, pcm, opus, aac, flac (provider support varies) |
--speed [speed] | - | Playback speed multiplier (OpenAI: 0.25-4.0, Cartesia: 0.6-1.5) |
--instructions [text] | - | Style instructions for speech (only gpt-4o-mini-tts and ElevenLabs). E.g. "Speak in a calm, soothing tone" |
--language [lang] | - | ISO 639-1 language code (e.g. en, es, fr). ElevenLabs and Cartesia |
--stdin | - | Read text from stdin instead of the positional argument |
--json | - | Output result metadata as JSON to stdout (model, file path, cost) |
--stdout | - | Write raw audio bytes to stdout instead of saving to a file. Useful for piping |
Global Options
| Option | Default | Description |
-h, --help | - | Display this message |
-v, --version | - | Display version number |
Examples
1# Generate speech with default model
1egaki speech "Hello, welcome to egaki!"
1# Use a specific model and voice
1egaki speech "Good morning" -m tts-1-hd --voice nova
1# Use GPT-4o Mini TTS with style instructions
1egaki speech "Breaking news today" -m gpt-4o-mini-tts --instructions "Speak like a news anchor"
1# ElevenLabs with a voice ID
1egaki speech "Hello world" -m eleven_v3 --voice 21m00Tcm4TlvDq8ikWAM
1# Read text from stdin
1cat script.txt | egaki speech --stdin -o narration.mp3
1# Cartesia Sonic 3.5 with a voice ID
1egaki speech "Hello world" -m sonic-3.5 --voice f786b574-daa5-4673-aa0c-cbe3e8534c02
1# Pipe audio to another tool
1egaki speech "test" --stdout | ffplay -nodisp -autoexit -