Biteable AI Functionality
The AI features we offer are designed to perform specific functions based solely on the information provided by users for each individual request. Below is a summary of the AI-enabled features we use and how they operate:
When users provide a prompt to generate a video, they input a title and a brief description of their use case (e.g., video title, business description). AVA submits a structured request to OpenAI to generate a script, which is then used in a video template. Keywords from the script help select relevant stock footage for the video. The data provided by the user is used strictly for generating the script and the video.
Biteable offers AI-powered voiceovers using Amazon Polly. Users can submit up to 1,000 characters of text to generate voiceover clips for their videos. The text is processed solely to generate the voiceover and cannot be created in bulk. Users must manually input the text each time they wish to generate speech.
Users who have created a video can request automated closed captions for any spoken audio. Biteable uses AWS Transcribe to generate these captions, which are fully editable. The audio data provided by the user is used only for the purpose of generating the captions.
When users add text to their videos, they have the option to translate the text into multiple supported languages. Biteable uses AWS Translate for this feature. Users must request translation for each text box individually, and the text is processed only for the requested translation.
All AI functionalities in Biteable require users to submit specific content (e.g., text, audio) to perform a task. This data is sent securely to OpenAI or AWS in a structured format (JSON) and returned immediately with the requested output (e.g., script, voiceover, translation, captions). Biteable does not store or analyze any of this data in aggregate, and it is not associated with user accounts. Data is processed only for the purpose of completing the specific task requested by the user and is not retained beyond that.
1. AVA (Automated Video Assistant)
When users provide a prompt to generate a video, they input a title and a brief description of their use case (e.g., video title, business description). AVA submits a structured request to OpenAI to generate a script, which is then used in a video template. Keywords from the script help select relevant stock footage for the video. The data provided by the user is used strictly for generating the script and the video.
2. AI Voices
Biteable offers AI-powered voiceovers using Amazon Polly. Users can submit up to 1,000 characters of text to generate voiceover clips for their videos. The text is processed solely to generate the voiceover and cannot be created in bulk. Users must manually input the text each time they wish to generate speech.
3. Automated Closed Captions
Users who have created a video can request automated closed captions for any spoken audio. Biteable uses AWS Transcribe to generate these captions, which are fully editable. The audio data provided by the user is used only for the purpose of generating the captions.
4. Text Translation
When users add text to their videos, they have the option to translate the text into multiple supported languages. Biteable uses AWS Translate for this feature. Users must request translation for each text box individually, and the text is processed only for the requested translation.
How We Handle Data
All AI functionalities in Biteable require users to submit specific content (e.g., text, audio) to perform a task. This data is sent securely to OpenAI or AWS in a structured format (JSON) and returned immediately with the requested output (e.g., script, voiceover, translation, captions). Biteable does not store or analyze any of this data in aggregate, and it is not associated with user accounts. Data is processed only for the purpose of completing the specific task requested by the user and is not retained beyond that.
Updated on: 24/09/2024
Thank you!