Meta has announced a demonstration of sound production using AI-powered tools called Audiobox, which allow you to create custom audio samples based on your voice or text requests.
Following the company’s preview of the new generative AI project Audiobox last month, which can mimic your voice using just a few seconds of audio sample.
Meta stated: “Audiobox represents our new experimental model for sound production.”
Audiobox can produce voices and sound effects by using a variety of audio inputs and natural Arabic language text prompts, making it easy to create distinctive sound for a wide range of different scenarios.
The interactive demonstration of the Audiobox model provides access to a variety of elements such as audio descriptions, sound effect creation, sound editing, and more.
The primary use case is to generate unique sound based on textual instructions.
Meta has provided some key elements to test Audiobox, including text-to-speech conversion, enabling you to create custom sound based on any textual input.
Meta is committed to offering two voice options, Alice and Emily, for use in the demonstration. This allows you to understand the capabilities of the process in translating text into alternative audio streams.
You can also add custom voices to your sample, aligning with textual requirements.
There is an option to create a unique sound for you, with extremely excellent and reliable results, which could raise significant concerns about potential misuse. However, Meta asks for your consent to a set of terms and conditions before trying it out.
In the increasingly popular list of AI generative tools, Audiobox emerges as a new and innovative model, set to have diverse applications in the coming year.
AI advancements are moving fast, and Meta aims to keep up with this progress. Meta continues to develop security standards, although there is concern about these tools being widely available, even with the attached conditions.
