What Bozos TTS does
Run /join while you are in a voice channel. Bozos TTS joins that channel and reads eligible chat messages aloud. /leave disconnects the TTS instance from your current voice channel without ending other active Bozos TTS sessions in the same server.
75 language families, 142 regional locales
The production voice catalog contains 75 language families across 142 verified regional locales. Popular choices include English, Hindi, Spanish, French, German, Portuguese, Arabic, Chinese, Japanese, Korean, Bengali, Tamil, Telugu, Marathi, Gujarati, Urdu, Indonesian, Turkish, Vietnamese and many more.
/language lets you choose either the Voice channel default or Only me. Personal language and accent preferences override the voice-channel default for that member.
323 verified neural voices
The current static production catalog contains 323 verified neural voices. After selecting a language and regional locale, use /voice to choose a compatible voice for the current voice channel or only for yourself.
Private voice previews
/voicepreview opens a private preview panel for the language/accent currently effective for you. Preview audio is sent as a private Discord voice message rather than being played through the live VC queue.
Grand Pronunciation Engine
Before speech is generated, Bozos applies a built-in pronunciation layer designed to make chat sound less robotic. It handles common slang, acronyms, gaming and technology terms, file formats, units, expressive spelling and Discord-style usernames.
Natural chat cleanup
The speech pipeline simplifies markdown, converts mentions into readable names, humanizes URLs, handles emoji, and normalizes punctuation, numbers, dates, times and other chat-oriented text patterns before synthesis.
Two speaker identity systems
The default Username System intelligently announces names only when they are useful. Consecutive messages from the same speaker avoid unnecessary repetition, and very small calls can suppress speaker names further.
/speaker-system set can switch the current voice channel to Spatial Speakers. In Spatial Speakers mode, usernames are not spoken; each member gets a stable spatial position plus a compatible voice. A member's personal /voice choice still takes priority. /speaker-system map shows the current assignments and /speaker-system test plays a Left → Center → Right headphone test.
Persistent Auto-Join
A member with Manage Server can use /auto-join enabled:true while Bozos is already connected to the same voice channel. Auto-Join remembers that VC and can reconnect Bozos automatically when its conditions are met.
Enabling Auto-Join requires Bozos to have View Channel, Connect, Speak and Send Messages in that voice channel. At least one person in the configured VC must also be a member of Bozos | Official Community for Auto-Join to operate. Disabling Auto-Join does not terminate an already-active live session.
Multi-VC support
Bozos TTS can coordinate the primary bot plus two voice-only worker identities. When those worker bots are available and installed in the server, up to three voice channels can have independent Bozos TTS sessions at the same time. The primary bot remains the control brain; the workers do not register slash commands or read message content.
Playback controls
/volumechanges playback from 0% to 200% for the current voice channel and requires Manage Server. The new gain applies immediately even to speech already playing./skipskips the message currently being spoken in your active TTS voice channel.- Queued messages can be pre-generated to reduce gaps between messages.
- Voice sessions preserve their queue through reconnect attempts when possible and clean themselves up automatically when appropriate.
Localized join/leave announcements
/announce lets a member with Manage Server enable or disable join/leave announcements for the current voice channel. They are off by default and use the TTS localization system.