tts Command¶

Generate speech from a text file using ElevenLabs TTS.

Synopsis¶

elevenlabs tts [flags] <text-file>

Description¶

The tts command converts a text file to speech using the ElevenLabs API. It supports:

Plain text input
SSML <break> tags for pauses
Emotion tags for the v3 model ([calm], [excited], etc.)
CAPITALIZED words for emphasis
YAML configuration files for reusable settings
Built-in presets for common use cases

Flags¶

Required¶

Flag	Short	Description
`--voice`	`-v`	Voice ID (required unless specified in config)

Output¶

Flag	Short	Default	Description
`--output`	`-o`	`<input>.mp3`	Output file path
`--format`	`-f`	`mp3_44100_128`	Audio format
`--model`	`-m`	`eleven_v3`	Model ID

Voice Settings¶

Flag	Range	Default	Description
`--stability`	0.0-1.0	0.5	Voice consistency (lower = more expressive)
`--similarity`	0.0-1.0	0.75	Adherence to original voice
`--style`	0.0-1.0	0.0	Style exaggeration
`--speed`	0.25-4.0	1.0	Speech rate multiplier

Configuration¶

Flag	Short	Description
`--config`	`-c`	Load settings from YAML config file
`--preset`	`-p`	Use preset: `oratory`, `podcast`, `audiobook`
`--save-config`		Save current settings to YAML file

Utility¶

Flag	Description
`--estimate`	Estimate credits without calling API

Output Formats¶

Format	Quality	Use Case
`mp3_22050_32`	Low	Small files, previews
`mp3_44100_128`	Good	General use (default)
`mp3_44100_192`	High	High-quality MP3
`pcm_44100`	Lossless	CD quality
`pcm_48000`	Lossless	Studio quality (recommended)
`opus_48000_128`	Efficient	Streaming, web
`ulaw_8000`	Telephony	Phone systems

Presets¶

Oratory¶

Optimized for speeches, presentations, and formal content:

stability: 0.4      # More expressive
style: 0.3          # Moderate dramatic emphasis
speed: 0.95         # Slightly slower for gravitas
format: pcm_48000   # Studio quality

Podcast¶

Optimized for conversational, natural delivery:

stability: 0.5      # Balanced
style: 0.0          # Natural, no exaggeration
speed: 1.0          # Normal pace
format: mp3_44100_128

Audiobook¶

Optimized for long-form narration:

stability: 0.6      # Consistent for long content
style: 0.1          # Subtle character
speed: 0.95         # Comfortable listening pace
format: pcm_48000   # Studio quality

Examples¶

Basic Usage¶

# Simple text-to-speech
elevenlabs tts -v IT8nQhZJj9jzRwmC46Ko speech.txt

# Specify output file
elevenlabs tts -v IT8nQhZJj9jzRwmC46Ko -o output.mp3 speech.txt

Using Presets¶

# Oratory preset for speeches
elevenlabs tts -v IT8nQhZJj9jzRwmC46Ko --preset oratory speech.txt

# Podcast preset
elevenlabs tts -v IT8nQhZJj9jzRwmC46Ko --preset podcast episode.txt

Custom Voice Settings¶

# More expressive with slower pace
elevenlabs tts -v IT8nQhZJj9jzRwmC46Ko \
  --stability 0.3 \
  --style 0.4 \
  --speed 0.9 \
  speech.txt

Configuration Files¶

# Use config file
elevenlabs tts --config tts-config.yaml speech.txt

# Save current settings to config
elevenlabs tts -v IT8nQhZJj9jzRwmC46Ko --preset oratory \
  --save-config my-config.yaml speech.txt

# Config overrides preset, CLI flags override config
elevenlabs tts --preset podcast --config custom.yaml --speed 1.1 speech.txt

High-Quality Output¶

# Studio quality PCM (48kHz, 16-bit, mono)
elevenlabs tts -v IT8nQhZJj9jzRwmC46Ko -f pcm_48000 speech.txt
# Output: speech.wav

Credit Estimation¶

# Estimate credits before generating (no API call)
elevenlabs tts -v IT8nQhZJj9jzRwmC46Ko --estimate speech.txt

# With config file
elevenlabs tts --config tts-config.yaml --estimate speech.txt

# Output:
# INFO credit estimate input=speech.txt words=453 characters=4600
#      speed=0.95 estimated_duration="3m 10s" estimated_credits=3178

The estimator calculates:

Words: Excluding SSML <break> tags and [emotion] markers
Duration: Based on ~150 WPM, adjusted for speed setting
Credits: ~1,000 credits per minute of audio

Input File Format¶

Plain Text¶

Hello, this is a simple text file.
It will be converted to speech.

With SSML Breaks¶

Hello. <break time="0.5s"/> This is a pause.
<break time="1s"/>
A longer pause before this sentence.

With Emotion Tags (v3 model)¶

[calm] Welcome to our presentation.
[excited] This is amazing news!
[firm] We must act now.

With Emphasis (Capitalization)¶

This is VERY important.
We are NO LONGER alone in intelligence.

Combined Example¶

[calm] <break time="1s"/>
There are moments <break time="0.15s"/>
in history <break time="0.2s"/>
when humanity does not merely advance —
it TRANSFORMS.
<break time="1s"/>

[serious] And so we must ask:
What makes us INDISPENSABLE?

Priority Order¶

Settings are applied in this order (later overrides earlier):

Defaults — Built-in default values
Preset — --preset flag values
Config file — --config file values
CLI flags — Individual flag values