Skip to content
English
Open Studio
  • There are no suggestions because the search field is empty.

Voice Cloning Process

Murf offers two voice cloning workflows: Instant Voice Cloning through the API and Professional Voice Cloning through a managed process. The recording requirements and creation timelines differ depending on which option you choose.

Instant Voice Cloning

Instant Voice Cloning allows you to create a custom voice from a single short audio sample using the Murf API.

Audio requirements

Prepare a reference recording that meets the following requirements:

  • Supported formats: WAV, MP3, FLAC, ALAW, and ULAW.
  • Maximum duration: 30 seconds.
  • Minimum sample rate: 24 kHz.
  • Maximum file size: 40 MB.
  • Recording quality: Clear, continuous speech from a single speaker, with minimal background noise.

If you provide a URL instead of uploading a file, it must be publicly accessible over HTTP or HTTPS and include a supported audio file extension. Private URLs and localhost URLs are not accepted.

If your source is a video or another unsupported file type, extract the audio and export it in a supported format before submitting it.

How to create a voice clone

  1. Request access: Instant Voice Cloning is available only on the Enterprise plan and must be enabled for your workspace. Contact your Customer Success Manager if you are already an Enterprise customer.
  2. Prepare your API key: Generate an API key from the Murf API Dashboard.
  3. Submit the reference audio: Call the Create Voice Clone endpoint with your audio file or a publicly accessible audio URL. Include a unique tag for the clone. You can also provide an optional displayName.
  4. Check the creation status: Use the request ID returned by the creation endpoint to check the job status.
  5. Retrieve the voice ID: When the status is COMPLETED, retrieve the generated voice ID.
  6. Generate speech: Use the voice ID with a supported Falcon 2.1 endpoint.

Voice creation runs asynchronously, and clones are generally ready within a few minutes. Always check the creation status rather than assuming the clone is ready after a fixed amount of time.

Professional Voice Cloning

Professional Voice Cloning follows a separate, managed process for customers who require a high-quality custom voice.

The process typically involves:

  1. Contacting the Murf Enterprise team to discuss your requirements.
  2. Preparing up to 90 minutes of studio-grade voice recordings.
  3. Working with the Murf team through the voice creation process.
  4. Using the completed voice through the supported Murf Studio or API workflow.

Professional Voice Cloning typically takes 1–4 weeks.

Important considerations:

  • Only clone a voice you own or have clear, documented consent to clone.
  • Instant Voice Cloning supports Falcon 2.1 only.
  • If a creation job fails, check the errorMessage returned by the status endpoint and verify the audio requirements.
  • Clones cannot be updated or retrained after creation. If you need a different result, create a new clone using a suitable reference recording.

Learn more

For endpoint details, request examples, status codes, and troubleshooting guidance, see the Instant Voice Cloning API Documentation.
Professional Voice Cloning Documentation