ElevenLabs for Home Assistant (built-in integration)

About

ElevenLabs is a built-in integration that adds text-to-speech and speech-to-text from the ElevenLabs cloud service. It provides one text-to-speech service and one speech-to-text service that Home Assistant can use for voice features. Setup happens in the UI and asks for an ElevenLabs API key. The options let the speech model and the voice be chosen, and the speech-to-text model can be picked separately. Speech-to-text can detect the spoken language automatically. Voice tuning covers similarity, stability and style, plus a speaker boost switch that raises similarity to the original voice.

Automatic summary of Home Assistant's source code. Something wrong? Get in touch

Supports

Uses the vendor's cloud Needs an internet connection and an account with the vendor.
Set up in the UI Add it in Settings, Devices and services. Nothing to download.
Updated with Home Assistant Ships and updates with every Home Assistant release.

Setting it up

It is added under Settings, Devices and services, with Add integration.

Connect

API keyYour ElevenLabs API key.

After setup

Configure voice settingsConfigure voice settings. Find more information in the ElevenLabs documentation.
ModelElevenLabs model to use. Please note that not all models support all languages equally well.
Auto-detect languageAutomatically detect the spoken language for speech-to-text.
Speech-to-text modelSpeech-to-text model to use.
VoiceVoice to use for text-to-speech.
SimilaritySimilarity of the generated audio to the original voice. Higher values may result in more similar audio, but may also introduce background noise.
StabilityStability of the generated audio. Higher values lead to less emotional audio.
StyleStyle of the generated audio. Recommended to keep at 0 for most almost all use cases.
Speaker boostUse speaker boost to increase the similarity of the generated audio to the original voice.