nuTTS fetches text-to-speech audio from free services such as lazypy.ro, Uberduck.ai, and a few services that offer TikTok TTS voices. By default, the resulting audio is returned in base64 encoded format wrapped in JSON. TTS audio can also be played using phiola. nuTTS is intended to be used with Twitch bots such as Firebot to provide TTS for alerts.

Basic Usage

Gets TTS audio for given service, returns base64 encoded audio

Results are output in JSON format

Usage:
  > nuTTS {flags} <text>

Subcommands:
  nuTTS config (custom) - Configures default settings for nuTTS
  nuTTS list (custom) - Gets JSON list of TTS voices
  nuTTS play (custom) - Plays TTS audio using phiola, returns playback status

Flags:
  -s, --service <string>: service to use for TTS voice (default: Streamlabs)
  -v, --voice <string>: voice ID for TTS (default: Brian)
  -h, --help: Display the help message for this command

Parameters:
  text <string>: text to speek

Example:

nuTTS --service Streamlabs --voice Emma "test message"

{
  "audioUrl": "data:audio/mp3;base64,<base64 encoded audio here>",
  "success": true,
  "audio_url": "https://lazypy.ro/tts/assets/audio/Streamlabs_Emma_c72b9698fa1927e1dd12d3cf26ed84b2.mp3",
  "info": "HTTP status: 200; Total transfer time: 0.333263 seconds.",
  "error_msg": null,
  "service_response": "{\"success\":true,\"message\":\"OK\",\"speak_url\":\"https:\\/\\/polly.streamlabs.com\\/v1\\/speech?OutputFormat=mp3&Text=test%20message&VoiceId=Emma&Engine=standard&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAURHYCFGLCVRRFGR5%2F20241208%2Fus-west-2%2Fpolly%2Faws4_request&X-Amz-Date=20241208T050441Z&X-Amz-SignedHeaders=host&X-Amz-Expires=900&X-Amz-Signature=c2627c4f334b9f4e1460d44460713b1d794c8aee1e828019c64ba3dfc146548f\"}",
  "meta": {
    "service": "Streamlabs",
    "voice_id": "Emma",
    "voice_name": "",
    "text": "test message",
    "playlist_index": 0
  }
}

Configure nuTTS Settings

Configures default settings for nuTTS

If no flags are used, interactive configuration will be ran
Results are output in JSON format unless ran in interactive mode

Usage:
  > nuTTS config {flags}

Flags:
  -s, --service <string>: set default TTS service (must be used with --voice flag)
  -v, --voice <string>: set default TTS voice (must be used with --service flag)
  -d, --device <int>: set default playback device for 'play' subcommand
  -t, --timeout <int>: set default playback timeout for 'play' subcommand
  -V, --volume <int>: set default playback volume for 'play' subcommand
  -w, --wait <duration>: set default time to wait before playback for 'play' subcommand
  -h, --help: Display the help message for this command

Example:

nuTTS config --service Streamlabs --voice Emma --device 2 --timeout 20 --volume 10 --wait 0sec

{
  "service": "Streamlabs",
  "voice": "Emma",
  "device": 2,
  "timeout": 20,
  "volume": 10,
  "wait": "0sec"
}

TTS Voice List

Gets JSON list of TTS voices

Usage:
  > nuTTS list {flags}

Flags:
  -s, --service <string>: get voices for specific service
  -h, --help: Display the help message for this command

Example:

nuTTS list --service TikTok

{
  "charLimit": 300,
  "countBytes": true,
  "voices": [
    {
      "vid": "en_uk_001",
      "name": "Narrator (Chris)",
      "flag": "GB",
      "lang": "English",
      "accent": "England",
      "gender": "M"
    },
    {
      "vid": "rest_of_voices_here",
      "name": "Rest Of Voices Here",
      "flag": "GB",
      "lang": "English",
      "accent": "England",
      "gender": "M"
    }
  ]
}

Play TTS Audio With phiola

Plays TTS audio using phiola, returns playback status

Results are output in JSON format by default

Usage:
  > nuTTS play {flags} <text>

Flags:
  -s, --service <string>: service to use for TTS voice (default: Streamlabs)
  -v, --voice <string>: voice ID for TTS (default: Brian)
  -d, --device <int>: device number for TTS playback; 0 is system default (default: 0)
  -e, --exit-code: output phiola exit code instead of JSON (default: JSON)
  -t, --timeout <int>: max playback seconds for TTS (default: 60)
  -V, --volume <int>: playback volume for TTS (0-100) (default: 100)
  -w, --wait <duration>: seconds to wait before starting TTS playback (default: 0sec)
  -h, --help: Display the help message for this command

Parameters:
  text <string>: text to speek

Example:

nuTTS play --device 2 --volume 10 --timeout 20 --wait 5sec --service Streamlabs --voice Emma "test message"

{
  "stdout": "\n#1 \" - \" \"@stdin\" 0MB 0:20.000 (441,000 samples) 48kbps MPEG1-L3 float32 22050Hz mono\n\n[................................................................] 0:00 / 0:20\r[===.............................................................] 0:01 / 0:20\n",
  "stderr": "φphiola v2.2.8 (linux-amd64)\n",
  "exit_code": 0
}