This section provides a brief overview of popular tools. The tools are described for information purposes only and are not an endorsement. Please consult with your instructor, supervisor, and journal publication for their generative AI usage policies. Approach usability and application with caution and discretion.
Last Updated: March 2024
Broadly speaking, generative AI audio tools can be broken down into voice generation/cloning, music, and general audio categories. Generative Audio AI can generate melodies or rhythms, transfer the “style” of one instrument to another, and even search through huge databases of audio to create new sounds! Furthermore, there are many AI tools to assist with audio editing, which will be appearing in popular Digital Audio Workstation software. Some tools offer text-to-speech (TTS) capabilities, which convert text to spoken audio, synthesizing a realistic, rather than robotic, voice. This is an area of emerging importance as we enter a world with realistic human voice cloning.
These models can be run locally, or via huggingface. They do not have the polish of ElevenLabs, but can be fine-tuned using your own datasets.
ElevenLabs creates natural-sounding speech using text-to-speech (TTS) software.
It can convert content into another language and supports 29 languages as well as 50+ accents. With the AI voice cloning feature, you can create an AI version of your voice with a few minutes of audio. The AI Voice Changer tool converts your voice into another character. ElevenLabs can generate voices for videos, games, podcasts, audiobooks, and more.
Elevenlabs provides examples for their platform on their speech synthesis prompting webpage. Below are some prompts that can be used with Elevenlabs:
As with other tools, do not share or input personal or confidential information. Review Elevenlabs privacy policy page for updated information.
Trial access is available through the CITL Innovation Studio.
