Glossary
Wake word
A wake word is a fixed phrase, such as Hey Google or Hey Siri, that a voice assistant listens for on the device before it starts handling a request. Until the phrase is heard, the assistant is only checking for that phrase.
Updated October 17, 20264 cited facts, each traced to its source
Also called hotword, trigger word, wake phrase.
A wake word (also called a hotword, trigger word or wake phrase) is the phrase that switches a voice assistant from waiting to listening. Say "Hey Google" or "Hey Siri" and the device treats the speech that follows as a request. Until then, a small detector runs on the device and looks only for that phrase.
The detector runs on the device first
Wake-word detection is a separate, always-on step that runs before any request is handled. Apple's 2017 article on Hey Siri says the detector uses a Deep Neural Network to turn the acoustic pattern of your voice into probabilities over speech sounds 1.
Microsoft describes the same pattern for Copilot on Windows. Its PC detects the Hey Copilot wake word locally, using an on-device service that keeps a 10-second audio buffer in memory which is never recorded or stored 2. The buffer exists only so the phrase can be matched. It is not a recording of your day.
Wake words misfire
A detector that listens all the time will sometimes trigger on speech that is not meant for it. In a 2020 study, a Google Home Mini playing the TV show The West Wing averaged 0.95 activations per hour without the wake word, which the authors singled out as a notable case 3.
So a wake word trades one problem for another. It frees you from pressing a key, but the device spends its time listening. A false wake can also start a request you never made.
Why it matters for dictation
Dictation usually starts from a key you press, not a spoken phrase. On a Mac, pressing and releasing the Microphone key starts Dictation, while pressing and holding it activates Siri, if Siri is enabled 4. A key press is deliberate, so it does not trigger on the television, a colleague or a radio in the room.
For writing, that is an advantage. You decide exactly when the words begin, and nothing needs to be listening for a phrase in the background. If you want hands-free writing, a wake word is one option, but it adds a listening step that a key does not.
Related terms
- Voice assistant: the product that a wake word starts.
- Voice activity detection: detecting that someone is speaking, which is related to but different from detecting a specific phrase.
- Push-to-talk: starting input with a button instead of a phrase.
- Turn detection: deciding when a request has ended once it has started.
Guides on this topic
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.
- 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.
- 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.
Pages that use this term
- Speech to Text AI: Dictation, Assistants and Voice AgentsTopic
- Voice Access in Windows 11: Setup and How It WorksGuide
- Voice Assistants vs Dictation: Siri, Alexa, GeminiGuide
- Home Assistant Voice: Speech to Text and Voice CommandsGuide
- On-Device vs Cloud Speech Recognition: The TradeoffsGuide
- Can an ESP32 or microcontroller do speech recognition?Question
- Can you switch back from Gemini to Google Assistant?Question
- How do you control smart home devices with Google Assistant voice commands?Question
- How do you set up voice commands in Home Assistant?Question
- How do you use Voice Access in Windows 11?Question
- Is Gemini replacing Google Assistant?Question
- What is the best speech to text for Home Assistant?Question
- What is the Home Assistant Voice Preview Edition?Question
- What voice commands does Google Assistant understand?Question
- What voice commands does Home Assistant understand?Question
- Which Google Assistant features does Gemini not support?Question
- False accept (accidental activation)Glossary
- Voice assistantGlossary