{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://catalog.lintel.tools/schemas/schemastore/everyvoice-tts-toolkit-data-configuration/versions/0.3.json",
  "title": "PreprocessingConfig",
  "x-lintel": {
    "source": "https://raw.githubusercontent.com/EveryVoiceTTS/everyvoice/main/everyvoice/.schema/everyvoice-shared-data-0.3.json",
    "sourceSha256": "b2cf1f8c9e8809c449c939960b882932f316eb67c5bc8399122caaad11394900",
    "fileMatch": [
      "everyvoice-shared-data.yaml",
      "everyvoice-shared-data.json"
    ],
    "parsers": [
      "json",
      "yaml"
    ]
  },
  "type": "object",
  "properties": {
    "dataset": {
      "default": "YourDataSet",
      "description": "The name of the dataset.",
      "title": "Dataset",
      "type": "string"
    },
    "train_split": {
      "default": 0.9,
      "description": "The amount of the dataset to use for training. The rest will be used as validation. Hold some of the validation set out for a test set if you are performing experiments.",
      "maximum": 1.0,
      "minimum": 0.0,
      "title": "Train Split",
      "type": "number"
    },
    "dataset_split_seed": {
      "default": 1234,
      "description": "The seed to use when splitting the dataset into train and validation sets.",
      "title": "Dataset Split Seed",
      "type": "integer"
    },
    "save_dir": {
      "default": "preprocessed/YourDataSet",
      "description": "The directory to save preprocessed files to.",
      "format": "path",
      "title": "Save Dir",
      "type": "string"
    },
    "audio": {
      "allOf": [
        {
          "$ref": "#/$defs/AudioConfig"
        }
      ],
      "description": "Configuration settings for audio."
    },
    "path_to_audio_config_file": {
      "anyOf": [
        {
          "format": "file-path",
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "description": "The path to an audio configuration file.",
      "title": "Path To Audio Config File"
    },
    "source_data": {
      "description": "A list of datasets.",
      "items": {
        "$ref": "#/$defs/Dataset"
      },
      "title": "Source Data",
      "type": "array"
    }
  },
  "$defs": {
    "AudioConfig": {
      "$schema": "http://json-schema.org/draft-07/schema",
      "type": "object",
      "properties": {
        "min_audio_length": {
          "default": 0.4,
          "description": "The minimum length of an audio sample in seconds. Audio shorter than this will be ignored during preprocessing.",
          "title": "Min Audio Length",
          "type": "number"
        },
        "max_audio_length": {
          "default": 11.0,
          "description": "The maximum length of an audio sample in seconds. Audio longer than this will be ignored during preprocessing. Increasing the max_audio_length will result in larger memory usage. If you are running out of memory, consider lowering the max_audio_length.",
          "title": "Max Audio Length",
          "type": "number"
        },
        "max_wav_value": {
          "default": 32767.0,
          "description": "Advanced. The maximum value allowed to be in your wav files. For 16-bit audio, this should be (2**16)/2 - 1.",
          "title": "Max Wav Value",
          "type": "number"
        },
        "input_sampling_rate": {
          "default": 22050,
          "description": "The sampling rate describes the number of samples per second of audio. The 'input_sampling_rate' is with respect to your vocoder, or spec-to-wav model. This means that the spectrograms predicted by your text-to-spec model will also be calculated from audio at this sampling rate. If you change this value, your audio will automatically be re-sampled during preprocessing.",
          "title": "Input Sampling Rate",
          "type": "integer"
        },
        "output_sampling_rate": {
          "default": 22050,
          "description": "Advanced. The sampling rate describes the number of samples per second of audio. The 'output_sampling_rate' is with respect to your vocoder, or spec-to-wav model. This means that the wav files generated by your vocoder or spec-to-wav model will be at this sampling rate. If you change this value, you will also need to change the upsample rates in your vocoder. Your audio will automatically be re-sampled during preprocessing.",
          "title": "Output Sampling Rate",
          "type": "integer"
        },
        "alignment_sampling_rate": {
          "default": 22050,
          "description": "Advanced. The sampling rate describes the number of samples per second of audio. The 'alignment_sampling_rate' describes the sampling rate used when training an alignment model. If you change this value, your audio will automatically be re-sampled during preprocessing.",
          "title": "Alignment Sampling Rate",
          "type": "integer"
        },
        "target_bit_depth": {
          "default": 16,
          "description": "Advanced. This is the bit depth of each sample in your audio files.",
          "title": "Target Bit Depth",
          "type": "integer"
        },
        "n_fft": {
          "default": 1024,
          "description": "Advanced. This is the number of bins used by the Fast Fourier Transform (FFT).",
          "title": "FFT Size",
          "type": "integer"
        },
        "fft_window_size": {
          "default": 1024,
          "description": "Advanced. This is the window size used by the Fast Fourier Transform (FFT).",
          "title": "FFT Window Size",
          "type": "integer"
        },
        "fft_hop_size": {
          "default": 256,
          "description": "Advanced. This is the hop size for calculating the Short-Time Fourier Transform (STFT) which calculates a sequence of spectrograms from a single audio file. Another way of putting it is that the hop size is equal to the amount of non-intersecting samples from the audio in each spectrogram.",
          "title": "FFT Hop Size",
          "type": "integer"
        },
        "f_min": {
          "default": 0,
          "description": "Advanced. This is the minimum frequency for the lowest frequency bin when calculating the spectrogram.",
          "title": "Minimum Frequency",
          "type": "integer"
        },
        "f_max": {
          "default": 8000,
          "description": "Advanced. This is the maximum frequency for the highest frequency bin when calculating the spectrogram.",
          "title": "Maximum Frequency",
          "type": "integer"
        },
        "n_mels": {
          "default": 80,
          "description": "Advanced. This is the number of filters in the Mel-scale spaced filterbank.",
          "title": "Number of Mel bins",
          "type": "integer"
        },
        "spec_type": {
          "anyOf": [
            {
              "$ref": "#/$defs/AudioSpecTypeEnum"
            },
            {
              "type": "string"
            }
          ],
          "default": "mel-librosa",
          "description": "Advanced. Defines how to calculate the spectrogram. 'mel' uses the TorchAudio implementation for a Mel spectrogram. 'mel-librosa' uses Librosa's implementation. 'linear' calculates a non-Mel linear spectrogram and 'raw' calculates a complex-valued spectrogram. 'linear' and 'raw' are not currently supported by EveryVoice. We recommend using 'mel-librosa'.",
          "title": "Spec Type"
        },
        "vocoder_segment_size": {
          "default": 8192,
          "description": "Advanced. The vocoder, or spec-to-wav model is trained by sampling random fixed-size sections of the audio. This value specifies the number of samples in those sections.",
          "title": "Vocoder Segment Size",
          "type": "integer"
        }
      },
      "title": "AudioConfig",
      "additionalProperties": false
    },
    "AudioSpecTypeEnum": {
      "enum": [
        "mel",
        "mel-librosa",
        "linear",
        "raw"
      ],
      "title": "AudioSpecTypeEnum",
      "type": "string"
    },
    "Dataset": {
      "$schema": "http://json-schema.org/draft-07/schema",
      "type": "object",
      "properties": {
        "label": {
          "default": "YourDataSet",
          "description": "A label for the source of data",
          "title": "Label",
          "type": "string"
        },
        "permissions_obtained": {
          "default": false,
          "description": "An attestation that permission has been obtained to use this data. You may not use EveryVoice to build a TTS system with data that you do not have permission to use and there are serious possible consequences for doing so. Finding data online does not constitute permission. The speaker should be aware and consent to their data being used in this way.",
          "title": "Permissions Obtained",
          "type": "boolean"
        },
        "data_dir": {
          "default": "/please/create/a/path/to/your/dataset/data",
          "description": "The path to the directory with your audio files.",
          "format": "path",
          "title": "Data Dir",
          "type": "string"
        },
        "filelist": {
          "default": "/please/create/a/path/to/your/dataset/filelist",
          "description": "The path to your dataset's filelist.",
          "format": "path",
          "title": "Filelist",
          "type": "string"
        },
        "filelist_loader": {
          "description": "Advanced. The file-loader function to use to load your dataset's filelist.",
          "title": "Filelist Loader",
          "type": "string"
        },
        "sox_effects": {
          "default": [
            [
              "channels",
              "1"
            ]
          ],
          "description": "Advanced. A list of SoX effects to apply to your audio prior to preprocessing. Run python -c 'import torchaudio; print(torchaudio.sox_effects.effect_names())' to see a list of supported effects.",
          "items": {},
          "title": "Sox Effects",
          "type": "array"
        }
      },
      "title": "Dataset",
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}