VideoContentSearchOcrAsrFeature

AI Overview😉

  • The potential purpose of this module is to analyze and match the text extracted from videos (Optical Character Recognition or OCR) with the audio transcript of the video (Automatic Speech Recognition or ASR). This helps to improve the accuracy of video content search results by ensuring that the video's audio and visual content are in sync.
  • This module could impact search results by giving more weight to videos where the OCR text and ASR transcript closely match, indicating a higher likelihood of relevance to the search query. This could lead to more accurate search results, especially for videos with spoken content. Conversely, videos with poor OCR-ASR matching may be demoted in search results.
  • To be more favorable for this function, a website could ensure that its video content has high-quality OCR text and ASR transcripts, which are accurately synchronized with the video's audio and visual content. This could involve using advanced OCR and ASR technologies, as well as manual review and correction of the extracted text and transcripts. Additionally, websites could provide additional metadata, such as captions or subtitles, to further improve the accuracy of the OCR-ASR matching process.

Interesting Module? Vote 👇

Voting helps other researchers find interesting modules.

Current Votes: 0

GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchOcrAsrFeature (google_api_content_warehouse v0.4.0)

Attributes

  • minCharEditDistance (type: integer(), default: nil) - The minimum char edit distance between the normalized OCR text and candidate word strings taken from a time window around the OCR appearance.
  • minCharEditDistanceAsrText (type: String.t, default: nil) - The matched ASR candidate for minimum char edit distance.
  • minCharEditDistancePercent (type: number(), default: nil) - The min_char_edit_distance divided by the length of the OCR string.
  • ocrTextNormalizedForCharMatch (type: String.t, default: nil) - The normalized OCR text which was used to match the candidate.
  • ocrTextNormalizedForCharMatchLength (type: integer(), default: nil) - The length of the normalized OCR text.
  • pretriggerScore (type: number(), default: nil) - The score from the pretrigger model.
  • wordOverlapAsrText (type: String.t, default: nil) - The ASR text that was used for the word overlap calculation.
  • wordOverlapCount (type: integer(), default: nil) - The number of words found both in the OCR text and the ASR in a time window around OCR appearance.
  • wordOverlapPercent (type: number(), default: nil) - The word_overlap_count divided by the number of words in the OCR text.

Summary

Types

t()

Functions

decode(value, options)

Unwrap a decoded JSON object into its complex fields.

Types

Link to this type

t()

@type t() :: %GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchOcrAsrFeature{
  minCharEditDistance: integer() | nil,
  minCharEditDistanceAsrText: String.t() | nil,
  minCharEditDistancePercent: number() | nil,
  ocrTextNormalizedForCharMatch: String.t() | nil,
  ocrTextNormalizedForCharMatchLength: integer() | nil,
  pretriggerScore: number() | nil,
  wordOverlapAsrText: String.t() | nil,
  wordOverlapCount: integer() | nil,
  wordOverlapPercent: number() | nil
}

Functions

Link to this function

decode(value, options)

@spec decode(struct(), keyword()) :: struct()

Unwrap a decoded JSON object into its complex fields.