Voicepipe TTS reads your text chat aloud in voice channels, so members who can't talk can still be heard.

Run /join in a text channel from the voice channel you're in. Voicepipe TTS joins and reads new messages from that channel and from the voice channel's own chat, each in the voice its author picked or the server's default. /skip skips the message being read, and /leave stops.

A voice for every member

• 59 voices in 7 languages: 30 Japanese (25 VOICEVOX, 5 Kokoro), 16 English (US and UK), and Spanish, French, Hindi, Italian and Brazilian Portuguese. Each member picks a language and then a voice with /voice, with a speed from 0.5 to 2.0
• The server sets a default voice for everyone who hasn't picked one
• Every voice's credit is on the voices page, in /help, and in the message Voicepipe TTS posts when it starts reading

Read the way your server talks

• A server dictionary: teach it how to read a word, such as "w" read as 「わら」. Up to 1,000 words, changed by members with Manage Server or a role you pick
• Japanese in any voice: a message with Japanese is pronounced as Japanese whichever voice reads it, and emoji, symbols and scripts no voice speaks are skipped rather than spelled out
• Only members in the call are read (on by default), so nobody outside it can make the bot speak
• A prefix for messages that shouldn't be read, ; by default
• Up to 150 characters of each message, adjustable from 20
• Auto-join: pair a voice channel with a text channel, and reading starts when the first person joins

Up to three voice channels at once

Add the helpers, Voicepipe TTS 2 and Voicepipe TTS 3, from /settings, and one server reads in up to three voice channels at the same time. A helper only joins voice channels and speaks. Voicepipe TTS reads the messages and posts for it.

Private channels stay private

A channel is read only while everyone in the voice channel can see it. If someone in the call can't, reading pauses there and the channel says so once.

Built for busy evenings

Synthesis capacity is sized ahead of the number of subscribed servers, and servers on the free plan get only what subscribed servers leave. If the synthesis machines can't be reached, a fallback engine keeps reading, with VOICEVOX's voices read in one of Kokoro's Japanese voices until they're back.

Permissions and privacy

Four permissions and nothing else: View Channels, Send Messages, Connect, Speak. Voicepipe TTS uses the Message Content intent to read messages aloud. The text is turned into speech and never stored, and voice channels are never recorded.

Pricing

Free plan: every feature and every voice, up to 1,000 characters read a day, in one voice channel at a time.

Subscription: $2.99 per month, per server, with no daily limit and up to three voice channels at once. Each server gets one 14-day free trial the first time it subscribes. Payments are handled by Stripe.