Text-to-Speech

Overview

60db’s Text-to-Speech (TTS) API converts written text into natural-sounding speech using advanced AI models. Our TTS engine supports multiple voices, languages, and customization options.

Features

Multiple Voices

Choose from 50+ pre-built voices or create custom voices

Voice Customization

Adjust speed, stability, and similarity

High Quality

Crystal-clear audio with natural intonation

Multiple Formats

Support for MP3, WAV, OGG, and FLAC output formats

Basic Usage

JavaScript
Python

import { SixtyDBClient } from '60db-service';

const client = new SixtyDBClient('your-api-key');

const audio = await client.textToSpeech({
  text: 'Hello, world!',
  voice_id: 'default-voice',
  enhance: true,
  speed: 1.0
});

from sixtydb import SixtyDBClient

client = SixtyDBClient('your-api-key')

audio = client.text_to_speech(
    text='Hello, world!',
    voice_id='default-voice',
    enhance=True,
    speed=1.0
)

with open('output.mp3', 'wb') as f:
    f.write(audio)

Voice Parameters

Speed

Control the speaking rate of the generated audio:

const audio = await client.textToSpeech({
  text: 'This will be spoken faster',
  speed: 1.5  // Range: 0.5 to 2.0
});

0.5: Half speed (slow)
1.0: Normal speed (default)
2.0: Double speed (fast)

Words per Minute

Alternatively, set an explicit target pace in words per minute (optional):

const audio = await client.textToSpeech({
  text: 'This will be spoken at an explicit pace',
  wpm: 137  // Range: 60 to 300 (optional — voice default when omitted)
});

Range: 60 – 300
When omitted, the voice’s default WPM is used (see the wpm field returned by GET /voices); voices without a measured default fall back to 137 WPM

Stability

Control how expressive versus consistent the generated voice sounds:

const audio = await client.textToSpeech({
  text: 'More consistent delivery',
  stability: 50  // Range: 0 to 100 (default 50)
});

Lower values produce more expressive, varied output
Higher values produce more consistent, stable output

Emotion Tags

Direct the delivery from inside the text with inline tags. Emotion tags change how the following sentences sound; sound tags add a laugh, sigh or pause:

const audio = await client.textToSpeech({
  text: "[apologetic] I'm sorry for the wait. [calm] Your ticket is now resolved. [happy] [laugh] Have a great day!"
});

Not sure where to put them? Use Enhance to have tags written for you. See Emotion Tags for the full list, combining rules and the Enhance Text endpoint.

Similarity

Control how closely the output matches the source voice:

const audio = await client.textToSpeech({
  text: 'Close clone match',
  similarity: 75  // Range: 0 to 100 (default 75)
});

Audio Enhancement

Generated audio is polished by default. Send audio_enhance: false to skip it when speed matters more than polish:

const audio = await client.textToSpeech({
  text: 'Clean, polished audio',
  audio_enhance: true  // Default: true. The older name `enhance` still works
});

Audio enhancement is never applied to live voice agents. The WebSocket TTS API always runs without it to keep latency low, and ignores any audio_enhance you send.

Output Formats

Supported audio formats:

Format	Quality	File Size	Use Case
MP3	Good	Small	Web, mobile apps
WAV	Excellent	Large	Professional audio
OGG	Good	Small	Web streaming
FLAC	Lossless	Medium	High-quality archival

const audio = await client.textToSpeech({
  text: 'Hello, world!',
  output_format: 'wav'  // mp3, wav, ogg, flac
});

Best Practices

Text Formatting

Use proper punctuation for natural pauses
Break long texts into paragraphs
Use SSML tags for advanced control (coming soon)

Voice Selection

Test multiple voices for your use case
Consider accent and gender for your audience
Use custom voices for brand consistency

Performance

Cache frequently used audio
Batch requests when possible
Use appropriate audio format for your use case

Quality

Enable enhancement for production use
Use WAV format for highest quality
Test with different speed settings

Use Cases

Voice Assistants

// Voice assistant
async function speakResponse(text) {
  const audio = await client.textToSpeech({
    text: text,
    voice_id: 'assistant-voice',
    enhance: true
  });

  playAudio(audio);
}

Content Narration

// Generate audiobook chapter
const audio = await client.textToSpeech({
  text: chapterText,
  voice_id: 'narrator-voice',
  speed: 0.95,
  output_format: 'mp3'
});

saveToFile(`chapter-${chapterNum}.mp3`, audio);

Accessibility

// Make web content accessible
async function readAloud(element) {
  const text = element.textContent;
  const audio = await client.textToSpeech({
    text,
    voice_id: 'clear-voice',
    enhance: true
  });
  
  playAudio(audio);
}

API Reference

For detailed API documentation, see:

Text to Speech

Standard TTS endpoint

Get Started

SDKs

Core Features

Text-to-Speech

Overview

Features

Multiple Voices

Voice Customization

High Quality

Multiple Formats

Basic Usage

Voice Parameters

Speed

Words per Minute

Stability

Emotion Tags

Similarity

Audio Enhancement

Output Formats

Best Practices

Use Cases

Voice Assistants

Content Narration

Accessibility

API Reference

Text to Speech

​Overview

​Features

Multiple Voices

Voice Customization

High Quality

Multiple Formats

​Basic Usage

​Voice Parameters

​Speed

​Words per Minute

​Stability

​Emotion Tags

​Similarity

​Audio Enhancement

​Output Formats

​Best Practices

​Use Cases

​Voice Assistants

​Content Narration

​Accessibility

​API Reference

Text to Speech

Overview

Features

Basic Usage

Voice Parameters

Speed

Words per Minute

Stability

Emotion Tags

Similarity

Audio Enhancement

Output Formats

Best Practices

Use Cases

Voice Assistants

Content Narration

Accessibility

API Reference