Skip to content

Extension "HoundTTS"

This extension provides integration with the HoundTTS text-to-speech engine for use with your DCS World server. For more information, refer to the HoundTTS documentation.

Installation

HoundTTS can be installed as a single mod package or in combination with different TTS providers. The easiest way is to use /extension enable <HoundTTS> to install it. If you have the ModManager enabled, the HondTTS extension will then download and install HoundTTS automatically.

If you want to install additional TTS providers, you can use the ModManager to install them:

Piper

Run /mods download repo:https://github.com/uriba107/HoundTTS in your admin channel in Discord. Select HondTTS-piper-engine-windows.zip and after that HoundTTS-piper-voices-windows.zip. Use /mods install to install both mods into the respective servers.

Supertonic

Run /mods download repo:https://github.com/uriba107/HoundTTS in your admin channel in Discord. Select HondTTS-supertonic-engine-windows.zip. Use /mods install to install the mod into the respective servers.

Configuration

You can use /extension enable <HoundTTS> to enable the extension. This will add an entry in your nodes.yaml like so:

# config/nodes.yaml
MyNode:
  # [...]
  instances:
    DCS.dcs_serverrelease:
      # [...]
      extensions:
        HoundTTS: 
          enabled: true           # enable this extension (default: true)
          autoupdate: true        # automatically update to the latest HoundTTS version (default: true)
          debug: false            # enable debug logging (default: false)
          DEFAULT_PROVIDER: sapi  # See HoundTTS documentation
          DEFAULT_VOICE: ''       # See HoundTTS documentation
          DEFAULT_CULTURE: en-US  # See HoundTTS documentation
          DEFAULT_GENDER: female  # See HoundTTS documentation
          provider:               # See HoundTTS documentation, Piper example for reference
            Piper:
              path: /path/to/piper.exe    # Optional, path to the Piper executable. Default: not set
              voice_path: /path/to/voices # Optional, path to .onnx files. Default: not set
              threads: 4                  # Optional, number of used threads. Default: 4
              max_concurrent: 4           # Optional, maximum number of parallel piper synthesizers. Default: 4


Configuration reference

Key Type Description Example
enabled bool Whether the HoundTTS text-to-speech extension is enabled. true
autoupdate bool Automatically update HoundTTS when a new version is available. true
debug bool Write HoundTTS debug output to the bot log. false
DEFAULT_PROVIDER str Default text-to-speech provider used for spoken output. sapi
DEFAULT_VOICE str Default voice used by the selected provider. en-US-JennyNeural
DEFAULT_CULTURE str Default language/culture for synthesized speech. en-US
DEFAULT_GENDER str Default voice gender used by the selected provider. female
provider map Per-provider engine settings and credentials. Piper:
path: C:\HoundTTS\piper
voice_path: C:\HoundTTS\voices
threads: 4
max_concurrent: 4

provider

Key Type Description Example
provider.Piper map Piper local neural text-to-speech engine settings. path: C:\HoundTTS\piper
voice_path: C:\HoundTTS\voices
threads: 4
max_concurrent: 4
provider.Piper.path str Directory that holds the Piper executable. C:\HoundTTS\piper
provider.Piper.voice_path str Directory that holds the Piper .onnx voice files. C:\HoundTTS\voices
provider.Piper.threads int Number of CPU threads each Piper synthesizer may use. (range: min 1) 4
provider.Piper.max_concurrent int Maximum number of Piper synthesizers run in parallel. (range: min 1) 4
provider.Google map Google Cloud text-to-speech provider settings. credentials_file: C:\HoundTTS\gcloud.json
provider.Google.credentials_file str Path to your Google Cloud service account JSON file. C:\HoundTTS\gcloud.json
provider.Azure map Microsoft Azure text-to-speech provider settings. key: '***'
region: westeurope
provider.Azure.key str Azure Speech subscription key. ***
provider.Azure.region str Azure region that hosts the Speech service. westeurope
provider.ElevenLabs map ElevenLabs text-to-speech provider settings. api_key: '***'
model_id: eleven_turbo_v2
provider.ElevenLabs.api_key str ElevenLabs API key. ***
provider.ElevenLabs.model_id str ElevenLabs model used for text-to-speech synthesis. eleven_turbo_v2
provider.AWS map Amazon Polly text-to-speech provider settings. access_key_id: ''
secret_key: '
'
region: eu-west-1
engine: standard
provider.AWS.access_key_id str AWS access key id used for Amazon Polly. ***
provider.AWS.secret_key str AWS secret access key used for Amazon Polly. ***
provider.AWS.region str AWS region that hosts the Polly service. eu-west-1
provider.AWS.engine str Polly speech engine to use, standard or neural. standard
provider.OpenAI map OpenAI text-to-speech provider settings. api_key: '***'
endpoint: https://api.openai.com/
model: tts-1
chat_model: gpt-4o-mini
provider.OpenAI.api_key str OpenAI API key. ***
provider.OpenAI.endpoint url Base URL of the OpenAI-compatible API. https://api.openai.com/
provider.OpenAI.model str OpenAI text-to-speech model name. tts-1
provider.OpenAI.chat_model str OpenAI chat model used to preprocess the spoken text. gpt-4o-mini
provider.LibreTranslate map LibreTranslate provider used to translate spoken text. api_key: '***'
endpoint: http://localhost:5000
provider.LibreTranslate.api_key str LibreTranslate API key. ***
provider.LibreTranslate.endpoint url URL of the LibreTranslate server. http://localhost:5000
provider.Discord map Discord text-to-speech provider settings. bot_token: '***'
provider.Discord.bot_token str Discord bot token used for voice output. ***
provider.Supertonic map Supertonic neural text-to-speech engine settings. path: C:\HoundTTS\supertonic
model_path: C:\HoundTTS\supertonic\models
voice_style: default
voice_style_path: C:\HoundTTS\supertonic\styles
total_steps: 8
threads: 4
max_concurrent: 4
provider.Supertonic.path str Directory that holds the Supertonic installation. C:\HoundTTS\supertonic
provider.Supertonic.model_path str Directory that holds the Supertonic model files. C:\HoundTTS\supertonic\models
provider.Supertonic.voice_style str Name of the Supertonic voice style to use. default
provider.Supertonic.voice_style_path str Directory that holds the Supertonic voice style files. C:\HoundTTS\supertonic\styles
provider.Supertonic.total_steps int Number of diffusion steps per synthesis; higher is slower. (range: min 1) 8
provider.Supertonic.threads int Number of CPU threads each Supertonic synthesizer may use. (range: min 1) 4
provider.Supertonic.max_concurrent int Maximum number of Supertonic synthesizers run in parallel. (range: min 1) 4