> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.vapi.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.vapi.ai/_mcp/server.

1. **Artifact Logging**: You can now access call logs directly through the new `logUrl` property in the [`Artifact`](https://api.vapi.ai/api#:~:text=Artifact) schema, providing a direct URL to call logs for each workflow execution to aid in debugging and compliance.

2. **Azure Speech Transcriber Segmentation**: You can now fine-tune speech segmentation in [`AzureSpeechTranscriber`](https://api.vapi.ai/api#:~:text=AzureSpeechTranscriber) using `segmentationStrategy`, `segmentationMaximumTimeMs`, and `segmentationSilenceTimeoutMs` properties for better transcription control.

> **Warning**
>
> **Breaking Changes**
>
> The following changes may require updates to your existing integrations:
>
> * The `messagePlan` and `backgroundDenoisingEnabled` properties have been removed from [`Assistant`](https://api.vapi.ai/api#:~:text=Assistant), [`AssistantOverrides`](https://api.vapi.ai/api#:~:text=AssistantOverrides), [`CreateAssistantDTO`](https://api.vapi.ai/api#:~:text=CreateAssistantDTO), and [`UpdateAssistantDTO`](https://api.vapi.ai/api#:~:text=UpdateAssistantDTO). Use `backgroundSpeechDenoisingPlan` instead for background denoising configuration.
>
> * The `segmentationMaxTimeMs` property in [`AzureSpeechTranscriber`](https://api.vapi.ai/api#:~:text=AzureSpeechTranscriber) has been replaced by `segmentationMaximumTimeMs` for consistency.
>
> * The `variables` field in [`Artifact`](https://api.vapi.ai/api#:~:text=Artifact) and [`NodeArtifact`](https://api.vapi.ai/api#:~:text=NodeArtifact) has been replaced by `variableValues` to standardize how extracted data is stored and referenced.