What is the process involved in voice cloning?

Regarding your query, the process of creating an AI voice (Voice Cloning feature) from recorded samples has a separate process and is not self-serve at the moment.

Here is a brief on the process that we follow:

• Murf will have the script prepared and shared to record/read for a specified tonality.
• We would require ~1-2 hrs of recordings for a given voice style. The recording must be noise-free and should have been done in a professional studio environment.
• The output should be a WAV file (Sample Rate: 48kHz, Bits per sample:16).
• You can also create different styles for the same voice; however, for each such style, a separate script with its corresponding recording data would be required.
• Once all the recording data is shared with us, we would require anywhere between 1-4 weeks to create an AI Avatar. You will own the rights to the custom voice and can also use it for commercial purposes.
• Please refer to the Voice Cloning page and the FAQs for more information about the feature and the process.