Migration Guide for Play AI's TTS API | Murf API

With Play.ai’s acquisition by Meta and subsequent service disruption, many developers are looking for a reliable alternative for their text-to-speech streaming needs. Murf offers a powerful, feature-rich streaming TTS API that provides an excellent migration path for Play.ai users.

This guide will help you seamlessly transition from Play.ai’s streaming API to Murf’s streaming solution.

Why Choose Murf?

Unparalleled Pronunciation Accuracy

Over 99% pronunciation accuracy for natural, precise speech

Superior Quality

Advanced GEN2 model for natural, human-like speech

Global Scale

Over 150 voices across 21+ languages and accents

Developer-First

Comprehensive SDKs, detailed documentation, and responsive support

Quick Migration Overview

Feature	Play.ai	Murf	Migration Notes
Endpoint	`POST /api/v1/tts/stream`	`POST /v1/speech/stream`	Simple URL change
Authentication	`X-USER-ID` + `Authorization`	`api-key` header	Simplified auth
Voice Selection	S3 paths or voice names	Voice IDs or names	More intuitive naming
Audio Formats	mp3, mulaw, raw, wav, ogg, flac	MP3, WAV, PCM	Supports all essential formats
Advanced Features	Limited customization	pronunciation, styles, MultiNative Locale	More control options

Step-by-Step Migration

Get Your Murf API Key

Test Connectivity

Verify your API key works with a simple request: bash curl -X GET https://api.murf.ai/v1/speech/voices \ -H "api-key: YOUR_API_KEY"

API Endpoint Migration

Change the API endpoint to Murf’s.

Play.ai Endpoint:

POST https://api.play.ai/api/v1/tts/stream

Murf Endpoint:

POST https://api.murf.ai/v1/speech/stream

Authentication Changes

Play.ai uses X-USER-ID and Authorization headers for authentication. Murf uses api-key header. Make sure to update your code to use the new header.

1 const headers = {
2 'X-USER-ID': 'your-user-id',
3 'Authorization': 'Bearer YOUR_API_KEY',
4 'Content-Type': 'application/json'
5 };

Request Body Parameter Mapping

Change the request body parameters to match Murf’s parameters. Only text, and voiceId are the required parameters and all other parameters are optional.

Play.ai Parameter	Murf Parameter	Notes
`text`	`text`	Direct mapping, no need to change
`voice`	`voiceId`	Use a murf voice ID, you can choose from here
`output_format`	`format`	”MP3”, “WAV”, or “PCM”
`sample_rate`	`sampleRate`	8000, 24000, 44100, or 48000
`speed`	`rate`	Range: -50 to +50 (vs Play.ai’s decimal values)

Any other Play.ai parameter that was not listed above should be removed. To see the full list of parameters that you can use, please refer to the Murf Streaming API Reference.

Process the Response

Both Play.ai and Murf return the audio stream in the response body. You can use the same code to process the response.

Code Migration Examples

Python

Node.js/JavaScript

curl

1 import requests
2 
3 url = "https://api.play.ai/api/v1/tts/stream"
4 headers = {
5     "X-USER-ID": "your-user-id",
6     "Authorization": "Bearer YOUR_API_KEY",
7     "Content-Type": "application/json"
8 }
9 data = {
10     "text": "Hello, this is a test message.",
11     "voice": "s3://voice-cloning-zero-shot/path/to/voice",
12     "voice_engine": "Play3.0",
13     "output_format": "mp3",
14     "sample_rate": 44100,
15     "speed": 1.0
16 }
17 
18 response = requests.post(url, headers=headers, json=data, stream=True)
19 with open("audio.mp3", "wb") as f:
20     for chunk in response.iter_content(chunk_size=1024):
21         f.write(chunk)

Voice Migration Guide

Finding Equivalent Voices

Play.ai used complex S3 paths for voice selection. Murf uses intuitive voice names and IDs.

Play.ai Voice Examples:

s3://voice-cloning-zero-shot/d9ff78ba-d016-47f6-b0ef-dd630f59414e/female-cs/manifest.json

Murf Voice Examples:

"en-US-natalie"     // Female, US English
"en-UK-oliver"      // Male, UK English
"es-ES-sofia"       // Female, Spanish
"fr-FR-marcel"      // Male, French

Voice Library

Use our Voice Library to find the perfect voice for your needs.

To programmatically fetch the list of voices, use the List voices endpoint.

1 from murf import Murf
2 
3 client = Murf(api_key="YOUR_API_KEY")
4 voices = client.voices.list()
5 
6 for voice in voices:
7     print(f"{voice.voice_id} - {voice.display_name} ({voice.gender})")