Technical Documentation for "AI Voice" by ASKtoAI
AI Voice: Perfect Voice Generation with Artificial Intelligence
Introduction
AI Voice is a cutting-edge solution that leverages artificial intelligence to create realistic synthetic voices. With a wide range of customizable options, AI Voice allows for the generation of vocal recordings from written text, offering a choice between various preset voices or the cloning of an existing voice.
Preliminary Phase: Text Input
Enter the text in the designated field. The model performs best with detailed paragraphs.
Voice Selection
Standard Voice
Users can choose from a variety of preset voices, each with unique characteristics:
- Rachel, Drew, Clyde, ...
- 🎅 Santa Claus, Grace, ...
- Gigi, Freya, ...
Cloned Voice
For even more detailed customization, users can opt to clone an existing voice by following these steps:
-
Fill in the fields with:
- Desired voice name;
- Brief description of the voice;
- Upload 0 to 25 audio samples. Quality should be prioritized over quantity. Noise-free audio files are recommended.
- Upload files by dragging them into the designated area or by clicking to select them. Audio or video files up to 10 MB each can be uploaded.
- Use the Add Voice button to complete the operation.
Voice Settings Configuration
-
Stability:
- More variable - For a more expressive and varied speech;
- More stable - For greater consistency, suited for longer texts.
- Clarity and Similarity Enhancement:
- Low - Recommended to minimize background audio artifacts;
- High - Improves the overall clarity of the sound and fidelity to the original voice.
-
Style Exaggeration:
- None (faster) - For rapid generation without special emphases;
- Exaggerated - To highlight the speaking style of the uploaded audio.
- Voice Enhancement: Increases fidelity to the source voice, resulting in slower generation speed.
- Restore Defaults: To undo all changes and return to standard configurations.
Audio File Generation
After setting the desired configurations, the audio file generation can proceed by clicking on ASKtoAI. The system will process the request and provide the vocal output based on the provided text and selected settings.