chromium/third_party/coremltools/mlmodel/format/AudioFeaturePrint.proto

// Copyright (c) 2021, Apple Inc. All rights reserved.
//
// Use of this source code is governed by a BSD-3-clause license that can be
// found in LICENSE.txt or at https://opensource.org/licenses/BSD-3-Clause

syntax = "proto3";
option optimize_for = LITE_RUNTIME;

package CoreML.Specification.CoreMLModels;

/*
 * A model which takes an input audio and outputs array(s) of features
 * according to the specified feature types
 */
message AudioFeaturePrint {
  // Specific audio feature print types

  // Sound extracts features useful for identifying the predominant
  // sound in audio signal
  message Sound {
    enum SoundVersion {
      SOUND_VERSION_INVALID = 0;
      // VERSION_1 is available on iOS,tvOS 15.0+, macOS 12.0+
      // It uses a variable-length input audio sample vector and yields a 512
      // float feature vector
      SOUND_VERSION_1 = 1;
    }

    SoundVersion version = 1;
  }

  // Audio feature print type
  oneof AudioFeaturePrintType {
    Sound sound = 20;
  }
}