- Voice Fallback Plan Introduced: You can now enhance your assistant’s reliability by defining fallback voice providers using
assistant.voice.fallbackPlan.voices. This allows your assistant to switch to alternative voices or providers likeFallbackLMNTVoice,FallbackAzureVoice,FallbackNeetsVoice,FallbackTavusVoice,FallbackOpenAIVoice, and others if the primary voice provider fails.

Refer to the FallbackPlan schema for more information: https://api.vapi.ai/api
-
Language Selection for PlayHTVoice: The
languageproperty has been added toPlayHTVoice. You can now specify the desired language for speech synthesis usingassistant.voice.language. -
AssemblyAI Transcriber Available: You can now use AssemblyAI for transcribing by setting
Assistant.transcribertoAssemblyAITranscriber. This provides a new option for converting speech to text in your assistant. -
Updated OpenAI Model Support: The
gpt-4o-2024-11-20model has been added toOpenAIModel.modelandOpenAIModel.fallbackModels. You can now configure your assistant to use this latest OpenAI model. -
Removal of ‘fillerInjectionEnabled’ Property: The
fillerInjectionEnabledproperty has been removed from voice configurations likeLMNTVoice,AzureVoice, etc. You no longer need to include this property when configuring these voices.