openai-tts

OpenAI TTS Example

This example demonstrates how to use OpenAI's Text-to-Speech API with the goframe voice package.

Features Demonstrated

1. Basic Text-to-Speech - Simple audio generation
2. Voice Selection - All 6 available OpenAI voices
3. HD Model - Higher quality audio synthesis
4. Streaming - Reduced latency for longer content
5. Audio Formats - MP3, Opus, AAC, FLAC

Prerequisites

1. OpenAI API Key - Get it from OpenAI Platform
2. API Credits - TTS API costs $15 per 1M characters (tts-1) or $30 per 1M characters (tts-1-hd)

Running the Example

# Set your OpenAI API key
export OPENAI_API_KEY="sk-..."

# Run the example
go run main.go

Output

The example generates multiple audio files:

• openai_tts_output.mp3 - Basic example
• openai_voice_alloy.mp3 - Alloy voice (neutral, versatile)
• openai_voice_echo.mp3 - Echo voice (warm, engaging)
• openai_voice_fable.mp3 - Fable voice (expressive, dramatic)
• openai_voice_onyx.mp3 - Onyx voice (deep, authoritative)
• openai_voice_nova.mp3 - Nova voice (energetic, friendly)
• openai_voice_shimmer.mp3 - Shimmer voice (soft, gentle)
• openai_hd_output.mp3 - HD quality example
• openai_streamed.mp3 - Streaming example
• openai_format_*.mp3 - Different audio formats

Voice Characteristics

Models

tts-1 (Standard)

• Faster generation
• Lower cost ($15 per 1M characters)
• Suitable for real-time applications
• Good quality for most use cases

tts-1-hd (High Definition)

• Higher quality audio
• Higher cost ($30 per 1M characters)
• Better for pre-recorded content
• Clearer pronunciation and prosody

Audio Formats

API Usage Examples

Basic Usage

synthesizer, _ := openai.NewSynthesizer(
    openai.WithAPIKey(os.Getenv("OPENAI_API_KEY")),
    openai.WithModel("tts-1"),
    openai.WithVoice("alloy"),
)

audio, _ := synthesizer.Synthesize(ctx, "Hello world!")
os.WriteFile("output.mp3", audio.Data, 0600)

With Options

audio, _ := synthesizer.Synthesize(ctx, 
    "Speaking faster now",
    voice.WithVoice("nova"),
    voice.WithModel("tts-1-hd"),
    voice.WithSpeed(1.2),
)

Streaming

stream, _ := synthesizer.Stream(ctx, "Long text...")
defer stream.Close()

file, _ := os.Create("output.mp3")
file.ReadFrom(stream)

Cost Considerations

• tts-1: $0.015 per 1K characters
• tts-1-hd: $0.030 per 1K characters
• Example: 10,000 characters (~2 min audio) = $0.15 (tts-1) or $0.30 (tts-1-hd)

Limitations

• Maximum text length: 4,096 characters per request
• Rate limits apply (check OpenAI dashboard)
• No word-level timestamps (use Kokoro-FastAPI for that)
• No subtitle generation built-in

Related Examples

• ../kokoro-tts/ - Local TTS with Kokoro-FastAPI
• ../kokoro-dialogue/ - Multi-speaker dialogue synthesis
• ../kokoro-captioned-dialogue/ - Dialogue with word-level timestamps