ttsrs - Text-to-Speech CLI Tool

A Rust-based command-line tool for converting text to speech using OpenAI's TTS API.

Table of Contents

• Features
• Installation
• Prerequisites
• Usage
  • Command-Line Arguments
  • Voice Options
  • Audio Formats
• Examples
• Environment Variables
• Technical Details

Features

• 🎯 Easy-to-use command-line interface
• 🔊 High-quality text-to-speech conversion using OpenAI's API
• 📝 Supports large text files through automatic chunking
• 🎨 Multiple voice options and audio formats
• 🔄 Interactive mode for selecting voices and formats
• 📁 Organized output with automatic file management
• 🚀 Progress indicators during conversion

Installation

cargo install ttsrs

Prerequisites

• Rust (latest stable version)
• ffmpeg (for audio file combining)
• OpenAI API key
• Internet connection

Usage

Command-Line Arguments

ttsrs [OPTIONS] <INPUT_FILE>

Argument	Description	Default
--model, -m	TTS model to use	tts-1-hd
--voice, -v	Voice selection	alloy
--format, -f	Output audio format	flac
--apikey, -a	OpenAI API key	-

Voice Options

Available voices with their characteristics:

• alloy - A versatile, well-balanced voice
• echo - Clear and professional, ideal for announcements
• fable - Warm and engaging, perfect for storytelling
• onyx - Deep and authoritative
• nova - Young and energetic
• shimmer - Soft and soothing

Audio Formats

Supported output formats:

• flac (default) - Lossless audio compression
• mp3 - Common compressed audio format
• wav - Uncompressed audio
• pcm - Raw audio data

Examples

Basic usage:

ttsrs input.txt

Specifying voice and format:

ttsrs --voice nova --format mp3 input.txt

Using API key inline:

ttsrs --apikey sk-... --voice echo --format wav input.txt

Environment Variables

• OPENAI_API_KEY: Your OpenAI API key

  export OPENAI_API_KEY='your-api-key-here'

Technical Details

• Text is automatically chunked into segments of 500 tokens or less
• Each chunk is processed separately and then combined
• Temporary files are automatically cleaned up
• Output is saved in a directory named after the input file

License

MIT License

Acknowledgments

Based on the unofficial-openai-tts-cli Python project.