AI Audio Plugins
Text-to-speech with ElevenLabs and EdgeTTS. Speech-to-text with Whisper.
ElevenLabs
High-quality, natural-sounding text-to-speech with multiple voices and languages.
Models
| Model | Credits | Best for |
|---|---|---|
eleven_v3 | 5 | Best quality |
eleven_flash_v2_5 | 2 | Low latency (default) |
eleven_multilingual_v2 | 3 | Multi-language |
eleven_turbo_v2_5 | 2 | Fast |
Popular voices
| Voice | ID | Character |
|---|---|---|
| Rachel | 21m00Tcm4TlvDq8ikWAM | Female, warm |
| Bella | EXAVITQu4vr4xnSDxMaL | Female, soft |
| Antoni | ErXwobaYiN019PkySvjV | Male |
| Josh | TxGEqnHWrfWFTfGW9XjX | Male, deep |
| Adam | pNInz6obpgDQGcFmaJgB | Male, narrator |
Config
{
"model_id": "eleven_flash_v2_5",
"voice_id": "21m00Tcm4TlvDq8ikWAM",
"stability": 0.5,
"similarity_boost": 0.75
}Ports
- Inputs:
text(text) - Outputs:
audio(audio)
EdgeTTS (Microsoft)
Free neural text-to-speech using Microsoft's Edge voices. 0-1 credits. Great for prototyping and budget workflows.
Whisper (OpenAI)
Speech-to-text transcription. Upload audio and get text back. 1-2 credits.
Ports
- Inputs:
audio(audio) - Outputs:
text(text)
Last updated: 2026-03-27