{
  "patent_number": "US 10847149",
  "country": "US",
  "title": "How Voice Assistants Listen for More Commands Without a Wake Word",
  "original_title": "Speech-based attention span for voice user interface",
  "summary": "This patent describes how a voice assistant can keep listening for follow-up commands after an initial request, without needing you to say its wake word again, based on the type of command you just gave.",
  "what_it_does": "The patent details a method for a voice assistant system to maintain an \"attention span\" after a user's initial command. After a device receives a first command, triggered by a keyword, and a server processes it and sends a response (Claim 1), the server decides if that command type (e.g., asking a question) is likely to be followed by another command. If so, the server instructs the device to continue sending audio data for a period without requiring the user to say the wake word again (Claim 1). For example, if you say \"Alexa, what's the weather?\" and it responds, the system might then listen for a follow-up like \"What about tomorrow?\" without you needing to say \"Alexa\" again. The system can also stop this \"attention span\" if the command type suggests it's no longer needed, such as after starting media playback (Claim 2).",
  "what_it_does_not_cover": [
    "Voice assistants that always require a wake word for every single command.",
    "Systems that keep listening indefinitely without a time threshold or specific command type trigger.",
    "Devices that decide to keep listening on their own, rather than being explicitly instructed by a server (Claim 1).",
    "Systems that only listen for follow-up commands if the user has explicitly enabled a general \"continuous listening\" mode, rather than based on the type of the previous command (Claim 1).",
    "Systems that process all subsequent audio as commands without first determining if it represents speech intended for processing (Claim 4)."
  ],
  "filed": "2017-09-01",
  "granted": "2020-11-24",
  "expires": "2037-09-01",
  "status": "active",
  "holder": "Amazon Technologies",
  "holder_url": "https://patentbrief.org/company/amazon-technologies",
  "inventors": [
    {
      "name": "Joseph Dean Nason Pemberton",
      "url": "https://patentbrief.org/inventor/joseph-dean-nason-pemberton"
    },
    {
      "name": "Eliav Samuel Zimmern Kahan",
      "url": "https://patentbrief.org/inventor/eliav-samuel-zimmern-kahan"
    },
    {
      "name": "Robert David Owen",
      "url": "https://patentbrief.org/inventor/robert-david-owen"
    },
    {
      "name": "Siu Ming Mok",
      "url": "https://patentbrief.org/inventor/siu-ming-mok"
    },
    {
      "name": "Diamond Bishop",
      "url": "https://patentbrief.org/inventor/diamond-bishop"
    }
  ],
  "times_cited": 87,
  "tags": [
    "consumer_electronics",
    "software",
    "telecommunications",
    "ai_ml"
  ],
  "abstract": "Techniques for enabling a device to send to a speech processing server further input audio data following a completed utterance dialog to prevent the need for subsequent keywords to be spoken to invoke subsequent commands are described. A system receives input audio data corresponding to an utterance from a device upon the device detecting speech corresponding to a keyword. The system performs speech processing on the input audio data to determine a command. The system determines output data responsive to the command and sends same to the device, thus completing operations regarding the utterance. The system may also send an instruction to the device to: send to the system further input audio data corresponding to further input audio without the device first detecting a wake command.",
  "url": "https://patentbrief.org/patent/us/10847149/speech-based-attention-span-for-voice-user-interface",
  "markdown_url": "https://patentbrief.org/patent/us/10847149/speech-based-attention-span-for-voice-user-interface/md",
  "google_patents_url": "https://patents.google.com/patent/US10847149",
  "relatedPatents": [
    {
      "patentNumber": "11361763",
      "countryCode": "US",
      "title": "How Smart Speakers Know You're Talking to Them After a Command",
      "url": "https://patentbrief.org/patent/us/11361763/detecting-system-directed-speech"
    },
    {
      "patentNumber": "10923111",
      "countryCode": "US",
      "title": "How Voice Assistants Distinguish Speakers After a Wakeword",
      "url": "https://patentbrief.org/patent/us/10923111/speech-detection-and-speech-recognition"
    },
    {
      "patentNumber": "10121471",
      "countryCode": "US",
      "title": "How Voice Assistants Know When You've Finished Talking",
      "url": "https://patentbrief.org/patent/us/10121471/language-model-speech-endpointing"
    },
    {
      "patentNumber": "11514901",
      "countryCode": "US",
      "title": "How Voice Assistants Recognize Who is Talking",
      "url": "https://patentbrief.org/patent/us/11514901/anchored-speech-detection-and-speech-recognition"
    },
    {
      "patentNumber": "9548050",
      "countryCode": "US",
      "title": "How a Digital Assistant Launches Apps Using Your Voice",
      "url": "https://patentbrief.org/patent/us/9548050/continuity-handoff"
    }
  ]
}