Text-to-Speech Audio Generation with Parallel AI

Text-to-Speech Audio Generation

Parallel AI integrates with ElevenLabs to offer high-quality text-to-speech audio generation. You can convert any text into natural-sounding audio using a wide range of voices — including public ElevenLabs voices and your own company-specific cloned voices. Use audio generation for:

Voiceovers for videos and presentations
Audio content for your voice agent channel
Automated announcements and notifications
Podcast-style content generation
Accessibility features in your product

Available Models

Retrieve the list of available audio models via a public endpoint — no authentication required.

curl "https://api.parallellabs.app/api/v0/audio/models"

The GET /api/v0/audio/models endpoint is public and does not require an API key. You can call it freely to discover available models before authenticating.

Available models include options across quality and speed tiers, such as:

Model ID	Description
`eleven_flash_v2_5`	Fast, low-latency model — ideal for real-time applications (default)
`eleven_multilingual_v2`	High-quality multilingual model supporting 29 languages
`eleven_turbo_v2`	Optimized for English with a balance of speed and quality

Available Voices

Your workspace has access to two categories of voices:

Public Voices

A large library of pre-built voices from ElevenLabs — diverse accents, tones, and styles for any use case.

Cloned Voices

Company-specific voices cloned from audio samples. These are private to your workspace and appear alongside public voices.

List all available voices for your workspace:

curl "https://api.parallellabs.app/api/v0/audio/voices" \
  -H "X-API-Key: YOUR_API_KEY"

Each voice object includes an id, name, labels (accent, gender, age, use case), and a preview URL. Use the id value as voiceId when creating audio.

Creating Audio

Send a POST /api/v0/audio request with your text and configuration to generate an audio file.

Request Parameters

Parameter	Type	Description
`text`	string	The text to convert to speech (required)
`voiceId`	string	ElevenLabs voice ID (default: `pNInz6obpgDQGcFmaJgB`)
`model`	string	Audio model to use (default: `eleven_flash_v2_5`)
`speed`	number	Playback speed multiplier (default: `1.0`)

curl -X POST "https://api.parallellabs.app/api/v0/audio" \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello, welcome to our platform!",
    "voiceId": "pNInz6obpgDQGcFmaJgB",
    "model": "eleven_flash_v2_5"
  }'

import requests

response = requests.post(
    "https://api.parallellabs.app/api/v0/audio",
    headers={
        "X-API-Key": "YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "text": "Hello, welcome to our platform!",
        "voiceId": "pNInz6obpgDQGcFmaJgB",
        "model": "eleven_flash_v2_5",
        "speed": 1.0
    }
)
print(response.json())

const response = await fetch(
  "https://api.parallellabs.app/api/v0/audio",
  {
    method: "POST",
    headers: {
      "X-API-Key": "YOUR_API_KEY",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      text: "Hello, welcome to our platform!",
      voiceId: "pNInz6obpgDQGcFmaJgB",
      model: "eleven_flash_v2_5",
      speed: 1.0,
    }),
  }
);
const data = await response.json();
console.log(data);

Retrieving Generated Audio

The create response includes an audioId. Use it to retrieve the audio file:

GET https://api.parallellabs.app/api/v0/audio/p/{audioId}

curl "https://api.parallellabs.app/api/v0/audio/p/{audioId}" \
  -H "X-API-Key: YOUR_API_KEY" \
  --output generated-audio.mp3

The /audio/p/{audioId} endpoint returns the raw audio binary. Pipe the output to a file (as shown above) or stream it directly to your application.

Managing Audio Files

Beyond creating and retrieving audio, you can list and delete your stored audio files.

List All Audio Files

curl "https://api.parallellabs.app/api/v0/audio?page=1&pageSize=20" \
  -H "X-API-Key: YOUR_API_KEY"

Delete an Audio File

curl -X DELETE "https://api.parallellabs.app/api/v0/audio/{audioId}" \
  -H "X-API-Key: YOUR_API_KEY"

Core Operations

Operation	Method	Endpoint
List audio files	`GET`	`/api/v0/audio`
Create audio	`POST`	`/api/v0/audio`
Retrieve audio file	`GET`	`/api/v0/audio/p/{audioId}`
Delete audio file	`DELETE`	`/api/v0/audio/{audioId}`
List voices	`GET`	`/api/v0/audio/voices`
List models	`GET`	`/api/v0/audio/models` (public)

Recommended Workflow

Discover available models

Call GET /api/v0/audio/models (no API key needed) to see the latest available models and choose one that fits your latency and quality requirements.

Browse voices

Call GET /api/v0/audio/voices to list all public and cloned voices available to your workspace. Note the id of the voice you want to use.

Generate audio

Send a POST /api/v0/audio request with your text, chosen voiceId, and model.

Retrieve the file

Use the returned audioId to fetch the audio binary from /api/v0/audio/p/{audioId} and save or stream it as needed.

Next Steps

Audio API Reference

Full reference for all audio endpoints including voices, models, and file management.

AI Agents

Deploy your agent on the voice channel to handle phone calls with natural-sounding audio.

Get Started

Core Features

Integrations

Platform

Text-to-Speech Audio Generation with Parallel AI