跳到主要内容

Audio Classification

POST 

/api/v5/audio/classification

Emotion classification and recognition of audio files

Request​

Query Parameters

    model stringrequired

    Model Name, default is audeering/wav2vec2-large-robust-12-ft-emotion-msp-dim

Header Parameters

    Authorization stringrequired

    User Personal Token

Body

    filebinaryrequired

    Audio file (support mp3, wav)

Responses​

Response Headers
    Schema
      vectornumber

      Vector representation of audio files

      emotion_class object
      arousalnumber

      Arousal Score (Emotional Activation Level: e.g., Excited/Calm)

      dominancenumber

      Dominance Score (Emotional Control: e.g., Dominant/Submissive)

      valencenumber

      Potency Score (Emotional Valence: e.g., Pleasant/Unpleasant)

      emotionItems object[]

      Emotion Classification Result List (Formatted)

    • Array [
    • labelstring

      Category Name

      Example: Excitement level
      labelEnstring

      Category Name (English)

      Example: Arousal
      valuestring

      Numbers

      Example: 0.6820112466812134
      valueDescstring

      Numerical Description

      Example: Excited
      valueDescEnstring

      Numerical Description (English)

      Example: slightly excited
    • ]