Glossary
Prompt (AI)
In AI, a prompt is the input you give a model to get a response, such as a question, an instruction or pasted text. Google's glossary defines it as any text entered as input to a large language model to condition the model to behave in a certain way.
Updated October 17, 20264 cited facts, each traced to its source
Also called prompt, voice prompt.
A prompt is the input you give an AI model so that it produces a response. For a chatbot, the prompt is the message you type or say. Google's Machine Learning Glossary defines a prompt as any text entered as input to a large language model to condition the model to behave in a certain way 1. The model's reply depends on that input, which is why the wording of a prompt changes the answer.
A prompt can be a question, an instruction or context
A prompt can be a short question, such as "What is a homophone?", or a long block of material. A typical working prompt has three parts: an instruction ("Summarize this email in three bullets"), the material to work on (the email itself) and any constraints (the length, the tone, what to leave out). Google's glossary notes that prompts can be as short as a phrase or arbitrarily long, such as the entire text of a novel.
Length has a measurable effect. A 2025 study found that longer prompts generally improved language model performance on domain-specific tasks, while shorter prompts could make it worse 2. The lesson is not to pad a prompt, but to include the background the model needs.
A voice prompt is a spoken prompt that is converted to text
A voice prompt is one you speak. Speech recognition turns it into text, and the model receives only that text. OpenAI describes its dictation feature this way: voice dictation is for turning speech into prompt text before you send it, and it is separate from ChatGPT Voice, which is a live spoken conversation 3. The distinction matters, because a spoken conversation and a dictated prompt have different goals.
Typed prompts can beat unedited voice prompts
A dictated prompt can carry transcription errors into the model's input, even when speaking feels quicker. In a 2026 preprint on programming students, those who typed their prompts were more likely than those who submitted unedited voice prompts to succeed on the first attempt, on two of three problems 4. The same study found no difference when students edited their transcribed voice prompts before submitting. Reading the transcript before you send it is the step that closes the gap.
Why it matters for dictation
When you dictate a prompt, the transcript is the prompt. A misheard name, a wrong number or a dropped negation changes what the model is asked to do, and the model will not know. Before you send, read the text, check the names and numbers, and make sure the instruction still says what you meant. Our guide to speech to text for AI explains how to speak prompts so they transcribe cleanly, and the speech to text and LLMs guide shows how the prompt fits into the wider pipeline.
Related terms
- Large language model (LLM): the model that a prompt is sent to.
- Generative AI: the class of models that respond to prompts by creating content.
- Speech language model: a model that can take a spoken prompt directly as audio.
- Machine learning: the broad method behind the models that prompts are sent to.
Guides that cover it
Evidence
Every number on this page, with where it comes from. Open a fact to see the exact words in its source.
- 1.
- 2.
- 3.
- 4.
Related terms
- Speech to Text AI: Dictation, Assistants and Voice AgentsThe topic this term belongs to
- AI text detectorAn AI text detector is software that estimates whether a passage was written by a language model, usually from the words alone. Detectors make mistakes in both directions, so a score is weak evidence about how a text was made.
- Generative AIGenerative AI is the class of AI models that create new content, such as text, audio or images, from patterns in their training data. The NIST profile defines it as models that emulate the structure and characteristics of input data to generate derived synthetic content.
- Generative error correction (GER)Generative error correction (GER) passes a speech recognizer's output, often its N-best list of alternative transcripts, to a large language model, which rewrites the transcript to fix errors. It can recover words the recognizer never offered, but it can also change words that were already right.
- Intent recognitionIntent recognition is the step that decides what a spoken request is asking for, such as setting a timer or sending a message, and which details the action needs. Voice assistants and voice agents rely on it before they act.
- Large language model (LLM)A large language model (LLM) is a language model with a very large number of parameters, trained on huge amounts of text to predict and generate language. Google's glossary defines the minimum as a language model having a very high number of parameters.
- Machine learningMachine learning is a method in which a program learns its behavior from example data rather than from rules written by hand. Google's glossary defines it as a program or system that trains a model from input data.
- Speech language modelA speech language model (speech LLM) is a large language model that accepts audio as input, so it can process speech directly instead of working only on a text transcript. The 2023 SALMONN model is one example that lets an LLM process general audio inputs.
- Speech-to-speech modelA speech-to-speech model takes audio in and produces audio out with one model, without writing the conversation out as text in between. Most voice systems instead chain speech recognition, a language model and speech synthesis.
- Text to speech (TTS)Text to speech (TTS), also called speech synthesis, is software that converts written text into spoken audio. Modern systems use neural networks, and they can run on the device or in the cloud.
- Turn detectionTurn detection is the step that decides when a speaker has finished a turn, so a voice system knows when to respond. A pause can be a hesitation rather than an ending, so the decision trades speed against cutting the speaker off.
- Voice agentA voice agent is a voice system that listens, reasons with a language model, uses tools, and speaks back across several turns to finish a task, such as checking an order and rescheduling it. Benchmarks show that agents still fail many tasks on realistic audio.
- Voice assistantA voice assistant is software that listens for a spoken request, works out what you want, and then acts or answers, such as setting a timer or turning off a light. Dictation is different: it turns your speech into text for you to edit.
Pages that use this term
- Speech to Text AI: Dictation, Assistants and Voice AgentsTopic
- AI Cleanup for Dictation: Fillers, Punctuation, RewritesGuide
- Voice Coding: How to Program by VoiceGuide
- The Future of Typing: Voice, AI and the KeyboardGuide
- Do longer prompts get better AI answers?Question
- How do you clean up dictated text with ChatGPT?Question
- Generative AIGlossary
- Large language model (LLM)Glossary