{
  "docId": "019dd923-5e88-73ef-bd5c-fd5012384ae3",
  "docSlug": "bf350dd574c19997",
  "documentTitle": "2021 Air Street Capital The State of AI Report 2021",
  "authorId": "AirStreetCapital",
  "authorName": "Air Street Capital",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "vc_research",
  "sourceTypeLabel": "VC research",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.777,
  "pageNumber": 35,
  "pageCount": 188,
  "prevPage": 34,
  "nextPage": 36,
  "slideType": "industry_trends",
  "function": "present_framework",
  "density": "dense",
  "nDataPoints": 0,
  "notes": "Includes a process diagram of the GSLM architecture.",
  "elementsJson": [
    "headline_text",
    "paragraph",
    "bullet_list",
    "process_diagram",
    "logo_grid"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5c-fd5012384ae3/35",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3#slide-35",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Generative Spoken Language Modeling (GSLM), the task of learning speech representations directly from raw audio without any labels or text.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c3-7407-8d26-d5325637c486",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.35,
        "w": 0.25,
        "x": 0.72,
        "y": 0.4
      },
      "kind": "diagram",
      "text": "GSLM Architecture: Quantizer Encoder (S2u) -> uLM -> Decoder (u2S) with Discrete Resynthesis",
      "attrs": null,
      "subkind": "process",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "57368448-2d7e-48ee-b9ec-01658a4a4499",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.35,
        "w": 0.68,
        "x": 0.02,
        "y": 0.38
      },
      "kind": "list",
      "text": "A major goal of GSLM is to make AI more inclusive: The majority of textual information available online is in a few languages like English. Better use of the audio information available online (podcasts, local radios, social apps) could help improve current AI audio systems' performance on rarer languages.\nThrough intonation, audio encodes more emotions and nuances. Being able to generate speech only from audio signals in a self-supervised fashion could result in more natural and expressive AI systems.\nThe researchers have already made some first steps in GSLM, by showing that they can leverage prosody (rhythm, stress and intonation of speech) to generate natural and coherent speech.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "590d8bc1-7cb8-47be-ac37-dbd38183c5a3",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.03,
        "w": 0.35,
        "x": 0.31,
        "y": 0.93
      },
      "kind": "paragraph",
      "text": "Some examples here: https://speechbot.github.io/pgslm/",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "a761bdf0-4a9c-44dc-a27b-b536640a0be8",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.12,
        "w": 0.68,
        "x": 0.02,
        "y": 0.21
      },
      "kind": "paragraph",
      "text": "Speech generation usually requires training an Automatic Speech Recognition (ASR) system, which is resource-intensive and error-prone. Researchers introduce Generative Spoken Language Modeling (GSLM), the task of learning speech representations directly from raw audio without any labels or text.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "c61cd26a-9234-489a-9286-fe1e7b38a766",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.04,
        "w": 0.45,
        "x": 0.02,
        "y": 0.14
      },
      "kind": "title",
      "text": "Beyond ASR for speech generation: textless NLP",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "a7459276-9e49-40af-abb1-75949837de1f",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "frameworks": [
    {
      "name": "process",
      "slug": null,
      "matchId": "05ac09da-d1d6-4a5a-8bef-293d4c00fd8e",
      "evidence": "The diagram illustrates the flow of data from audio through the GSLM components.",
      "confidence": 0.8
    }
  ],
  "arcBeats": [
    {
      "to": 152,
      "from": 5,
      "beatId": "019dd95a-0682-776c-8e35-0affbadec38e",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Definitions p.5-6, Exec Summary p.7, then Sections 1-3 catalog evidence.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 152,
      "from": 10,
      "beatId": "019dd95a-0682-776c-8e35-1b3ee3a1c012",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Three sections accumulating evidence of accelerating progress.",
      "position": 2,
      "confidence": 55,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}