by ElevenLabs
ElevenLabs provides AI voice tools: text to speech, speech to text, voice cloning, voice agents, music, sound effects and dubbing. You can use it through a no-code web app or through a REST API with Python and TypeScript SDKs.
Web appFreemium
Step 1: Choose a path: the web app for creative work, the agents platform, or the API.
Step 2: For the API, create an API key and set it as an environment variable.
Step 3: Install the Python or TypeScript SDK.
Step 4: Pick a voice from the voice library and a model, then send text to the text to speech endpoint.
Step 5: Save the returned audio to a file, or use the CLI to do the same from a terminal.
Step 6: Check usage in the analytics tab of the Developers dashboard.
Text to speech uses one credit per character of input text, and other operations are charged per second of audio processed. Credits reset monthly and unused credits roll over for up to two months.
Yes. You can clone a voice from an audio recording or generate one from a text description. Custom, cloned and designed voices can be referenced by ID through the API.
The default location is the U.S. Enterprise customers can choose data residency in the EU, India or Singapore, or run products inside their own cloud account with private deployments.
The docs give a median first-audio latency of about 100ms for the low-latency text to speech models and about 150ms for real-time transcription. These figures exclude application and network latency.
MurfMurf is an AI voice platform with a studio for voiceovers, an API for text to speech, and tools for dubbing and conversational voice agents. It offers 200+ voices across 35+ languages.
CleanvoiceCleanvoice is an AI editor that cleans podcast and video audio by removing filler words, silences, background noise and mouth sounds. It also has an API and SDK.
Murf AIMurf AI offers a suite of voice APIs: text to speech, dubbing, voice changer, translation and speech to text. It is aimed at developers building voice agents and localized audio.