Skip to content

Configuration File: confignew.json

erew123 edited this page Oct 11, 2024 · 3 revisions

The confignew.json file contains various configuration settings for the AllTalk application. Below is the structure of the file along with explanations for each section and setting. The base/default confignew.json can be downloaded from here https://github.com/erew123/alltalk_tts/blob/alltalkbeta/confignew.json

{
    "branding": "AllTalk ",
    "delete_output_wavs": "Disabled",
    "gradio_interface": true,
    "output_folder": "outputs",
    "gradio_port_number": 7852,
    "firstrun_model": true,
    "firstrun_splash": true,
    "launch_gradio": true,
    "transcode_audio_format": "Disabled",
    "theme": {
        "file": null,
        "class": "gradio/base"
    },
    "rvc_settings": {
        // RVC settings here
    },
    "tgwui": {
        // TGWUI settings here
    },
    "api_def": {
        // API settings here
    },
    "debugging": {
        // Debugging settings here
    }
}

Main Settings

  • branding: The name of the application displayed at the command prompt/terminal.
  • delete_output_wavs: Whether to delete output WAV files on start-up.
  • gradio_interface: Enable or disable the Gradio interface.
  • output_folder: The folder where TTS output files are saved.
  • gradio_port_number: The port number for the Gradio interface. API port number is in the API section (futher down).
  • firstrun_model: Flag for first run of the AllTalk to display the model choices on start-up.
  • firstrun_splash: Flag for displaying the first run splash screen in the Gradio interface.
  • launch_gradio: Whether to launch the Gradio interface on startup.
  • transcode_audio_format: Audio format for transcoding, or "Disabled" if not used.

Theme Settings

  • theme.file: Custom theme file (null if not used).
  • theme.class: The theme class to use (e.g., "gradio/base").

RVC Settings

The rvc_settings object contains configuration for RVC (Real-time Voice Conversion) as set in the Gradio interface:

"rvc_settings": {
    "rvc_enabled": false,
    "rvc_char_model_file": "Disabled",
    "rvc_narr_model_file": "Disabled",
    "split_audio": true,
    "autotune": false,
    "pitch": 0,
    "filter_radius": 3,
    "index_rate": 0.75,
    "rms_mix_rate": 1,
    "protect": 0.5,
    "hop_length": 130,
    "f0method": "fcpe",
    "embedder_model": "hubert",
    "training_data_size": 45000
}
  • rvc_enabled: Enable or disable RVC.
  • rvc_char_model_file: Character voice file for RVC.
  • rvc_narr_model_file: Narrator voice file for RVC.
  • split_audio: Whether to split audio while processing through RVC.
  • autotune: Enable or disable autotune.
  • pitch: Pitch adjustment.
  • filter_radius, index_rate, rms_mix_rate, protect, hop_length: Various RVC processing parameters.
  • f0method: F0 extraction method.
  • embedder_model: Model used for embedding.
  • training_data_size: Size of training data.

TGWUI Settings

The tgwui object contains settings for the Text Generation Web UI if AllTalk is installed as part of TGWUI:

"tgwui": {
    "tgwui_activate_tts": true,
    "tgwui_autoplay_tts": true,
    "tgwui_narrator_enabled": "false",
    "tgwui_non_quoted_text_is": "character",
    "tgwui_deepspeed_enabled": false,
    "tgwui_language": "English",
    "tgwui_lowvram_enabled": false,
    "tgwui_pitch_set": 0,
    "tgwui_temperature_set": 0.75,
    "tgwui_repetitionpenalty_set": 10,
    "tgwui_generationspeed_set": 1,
    "tgwui_narrator_voice": "female_01.wav",
    "tgwui_show_text": true,
    "tgwui_character_voice": "female_01.wav",
    "tgwui_rvc_char_voice": "Disabled",
    "tgwui_rvc_narr_voice": "Disabled"
}

These settings control various aspects of the Text Generation Web UI, including TTS activation, autoplay, narrator settings, language, and voice settings.

API Settings

The api_def object contains settings for the AllTalk API as set in the Gradio interface:

"api_def": {
    "api_port_number": 7851,
    "api_allowed_filter": "[^a-zA-Z0-9\\s.,;:!?\\-\\'\"$\\u0400-\\u04FF\\u00C0-\\u017F\\u0150\\u0151\\u0170\\u0171\\u011E\\u011F\\u0130\\u0131\\u0900-\\u097F\\u2018\\u2019\\u201C\\u201D\\u3001\\u3002\\u3040-\\u309F\\u30A0-\\u30FF\\u4E00-\\u9FFF\\u3400-\\u4DBF\\uF900-\\uFAFF\\u0600-\\u06FF\\u0750-\\u077F\\uFB50-\\uFDFF\\uFE70-\\uFEFF\\uAC00-\\uD7A3\\u1100-\\u11FF\\u3130-\\u318F\\uFF01\\uFF0c\\uFF1A\\uFF1B\\uFF1F]",
    "api_length_stripping": 3,
    "api_max_characters": 2000,
    "api_use_legacy_api": false,
    "api_legacy_ip_address": "127.0.0.1",
    "api_text_filtering": "standard",
    "api_narrator_enabled": "false",
    "api_text_not_inside": "character",
    "api_language": "en",
    "api_output_file_name": "myoutputfile",
    "api_output_file_timestamp": true,
    "api_autoplay": false,
    "api_autoplay_volume": 0.5
}

These settings control the API behavior, including port number, allowed characters, text filtering, language, and output file settings.

Debugging Settings

The debugging object contains flags for various debugging options:

"debugging": {
    "debug_transcode": false,
    "debug_tts": false,
    "debug_openai": false,
    "debug_concat": false,
    "debug_tts_variables": false,
    "debug_rvc": false
}

These flags can be set to true to enable debugging for specific components of the application.

Clone this wiki locally