Kitta AI Docs
Integration GuidesThird-Party Integrations
Back to site

AIRI Setup Guide

Configure Kitta AI as an OpenAI-compatible speech provider in AIRI and assign a cloned voice to an AI character.

Connect Kitta AI TTS to AIRI

AIRI has a dedicated OpenAI Compatible speech provider. Connect Kitta AI there, then select the provider, model, and Voice ID in AIRI's Speech module.

Prepare the API key and Voice ID

  1. Create a dedicated key in API management.
  2. Open your cloned voice and copy its Voice ID.
  3. Confirm that your API credit balance is sufficient.

For the first test, you can use this system voice:

00a1b221-6137-4b73-ad62-b0cbce134167

Add the speech provider

Open Settings → Providers → Speech → OpenAI Compatible, then enter:

SettingValue
Base URLhttps://kittaai.com/v1
API KeyYour Kitta AI API key
Modelfishaudio-s21pro-flash
Speed1

Use the API root, not the complete endpoint

AIRI's OpenAI-compatible provider appends /audio/speech. Use the Base URL ending at /v1; do not enter /v1/audio/speech in this field.

If automatic model discovery does not show the Kitta AI model, type fishaudio-s21pro-flash directly in the model field.

Select the voice

  1. Open Settings → Modules → Speech.
  2. Select the OpenAI Compatible provider and fishaudio-s21pro-flash.
  3. Enter the Kitta AI Voice ID as a custom voice.
  4. Enter a short sentence and click Test Voice.

After the system voice works, replace it with your cloned voice's Voice ID and assign it to the character or speech module you use.

Troubleshooting

  • 404: the Base URL should be https://kittaai.com/v1, without another /audio/speech suffix.
  • 401: the API key is invalid, revoked, or incomplete. Create a separate key for AIRI.
  • 402: add API credits before testing again.
  • Model is missing: enter fishaudio-s21pro-flash manually instead of relying on model discovery.
  • Voice list is empty: this is expected for a compatible provider; paste a Kitta AI Voice ID into the custom voice field.
  • No character audio: confirm the Kitta AI provider, model, and Voice ID are selected in the Speech module, not only saved on the Providers page.

LibreChat Setup Guide

Configure Kitta AI as an OpenAI-compatible server-side TTS provider in librechat.yaml and play replies with a cloned voice.

OpenTalking Setup Guide

Configure the OpenAI-compatible TTS provider in OpenTalking to use a Kitta AI model and cloned voice.

On this page

Connect Kitta AI TTS to AIRIPrepare the API key and Voice IDAdd the speech providerSelect the voiceTroubleshootingRelated documentation