AI voice generation is the use of generative AI to produce realistic human-sounding speech. It includes text-to-speech, voice cloning, voice conversion and AI dubbing, and is used for narration, assistants, accessibility and localisation.
AI video generation uses generative AI to create or modify video from text prompts, images, audio or existing footage. It includes text-to-video, image-to-video, AI avatars, video editing and automatic dubbing.
Structured output is when an AI model returns its response in a predefined, machine-readable format, such as JSON matching a schema, instead of free text, so software can reliably use the result.
Chain-of-thought (CoT) prompting is a technique that asks an AI model to work through a problem in intermediate steps before giving its final answer, which improves accuracy on maths, logic and multi-step tasks.
Few-shot prompting is including a small number of worked examples, showing an input and the desired output, in a prompt so the AI model learns the pattern and applies it to a new input.
Zero-shot prompting is giving an AI model a task using only instructions, without any examples of the desired output. The model relies on what it learned during training to understand and complete the task.
Prompt engineering is the practice of designing and refining the instructions, context and examples given to an AI model so it produces accurate, useful and consistent output.
AI hallucination is when a generative AI model produces information that sounds plausible and confident but is false, made up or unsupported by its sources, such as invented facts, quotes, citations or statistics.
Synthetic media is any image, audio, video or text that is generated or significantly altered by artificial intelligence. It includes AI art, synthetic voices and deepfakes, and brings both creative opportunities and serious risks of deception.
Multimodal AI is artificial intelligence that can understand, combine or generate more than one type of data, such as text, images, audio and video, for example answering questions about a photo or generating an image from text.
Speech-to-text (STT), also called automatic speech recognition (ASR), is technology that converts spoken audio into written text. It powers transcription, captions, voice assistants and call analytics.
Text-to-speech (TTS) is technology that converts written text into spoken audio. Modern neural TTS uses deep learning to produce natural-sounding speech with realistic rhythm, emphasis and tone.
Text-to-video is a generative AI technique that creates a short video clip from a written description. It extends image generation across time, so the model must keep objects, lighting and motion consistent from frame to frame.
Text-to-image is a generative AI technique that creates a new image from a written description. A text encoder turns the prompt into numbers, and an image model, usually a diffusion model, generates a picture that matches it.
AI music generation uses generative AI to compose melodies, create full songs with vocals, produce background tracks or generate sound effects from text prompts, audio examples or musical notation.