Eleven Labs Release Notes

Eleven v3 API

Eleven v3 is now available via the API.

To start using it, simply specify the model ID eleven_v3 when making Text to Speech requests.

Additionally the Text to Dialogue API endpoint is now available to all.

Music Generation API

The Eleven Music API is now freely available to all paid users.

Visit the quickstart to lean how to integrate. The API section below highlights the new endpoints that have been released.

Global TTS API preview

ElevenLabs is launching inference servers in additional geographical regions to reduce latency for clients outside of the US. Initial request processing will be available in the Netherlands and in Singapore in addition to the US.

To learn how to get started head to the docs.

API

Music

Eleven Music: Officially released new music generation model that creates studio-grade music with natural language prompts in any style. See the capabilities page and prompting guide for more information.

SDKs

v2.9.0 of the TypesScript SDK released

• Includes better typing support for Speech to Text requests in webhook mode
• Includes new enums for ChatGPT 5

v2.9.2 of the Python SDK released

• Includes new enums for ChatGPT 5

Agents Platform

Agent response correction: Updated WebSocket event schema and handling for improved agent response correction functionality.

API

Agents Platform

• Conversation token generation: Added new route to generate Conversation Tokens for WebRTC connections. Learn more
• Expandable widget options: Our embeddable widget can now be customized to start in the expanded state and disable collapsing altogether.
• Simplified operation IDs: We simplified the OpenAPI operator IDs for Agents Platform endpoints to improve developer experience.

Workspaces

• Simplified operation IDs: We simplified the operation IDs for our workspace endpoints to improve API usability.

SDK Releases

• Python SDK v2.8.2: Released latest version with improvements and bug fixes. View release

NPM Packages

• @elevenlabs/react-native@0.1.2: Enhanced React Native support
• @elevenlabs/client@0.4.4: Client library improvements
• @elevenlabs/react@0.4.5: React component updates

API

Workspaces

• Service account API key management: Added comprehensive API endpoints for managing service account API keys, including creation, retrieval, updating, and deletion capabilities. See Service Accounts documentation.

Agents Platform

• Post-call webhook migration: The post call webhook format is being migrated so that webhook handlers can be auto generated in the SDKs. This is not a breaking change, and no further action is required if your current handler accepts additional fields. Please see more information here.
• Agent transfer improvements: Fixed system variable system_agent_id to properly update after agent-to-agent transfers, ensuring accurate conversation context tracking. Added new system_current_agent_id variable for tracking current active agent. Learn more about dynamic variables.
• Enhanced public agent page: Added text input functionality and dynamic variable support to the public talk-to-agent page. You can now pass dynamic variables via URL parameters (e.g., ?var_username=value) and use text input during voice conversations. See dynamic variables guide. …

Agents Platform

• Azure OpenAI custom LLM support: Added support for Azure-hosted OpenAI models in custom LLM configurations. When using an Azure endpoint, a new required field for API version is now available in the UI.
• Genesys output variables: Added support for output variables when using Genesys integrations, enabling better call analytics and data collection.
• Gemini 2.5 Preview Models Deprecation: Modelsgemini-2.5-flash-preview-05-20 and gemini-2.5-flash-preview-04-17 have been deprecated in Agents Platform as they are being deprecated on 15th July by Google. All agents using these models will automatically be transferred to gemini-2.5-flash the next time they are used. No action is required.
• WebRTC rollout: Began progressive rollout of WebRTC capabilities for improved connection stability and performance. WebRTC mode can be selected in the React SDK and is used in 11.ai.
• Keypad touch tone: Fixed an issue affecting playing keypad touch tones on Twilio. See keypad touch tone documentation.

Voices …

Agents Platform

• HIPAA Compliance: Gemini 2.5 Flash is now available for HIPAA customers, providing enhanced AI capabilities while maintaining strict healthcare compliance standards.

• Post-call Audio: Added support for returning call audio in post-call webhooks, enabling comprehensive conversation analysis and quality assurance workflows.

• Enhanced Widget: Added additional text customization options including start chat button text, chatting status text, and input placeholders for text-only and new conversations.

• Agent Transfers: Improved agent transfer capabilities with transfer delay configuration, custom transfer messages, and control over transferred agent first message behavior.

• SIP Trunk Enhancements: Added support for separate inbound and outbound SIP trunk configurations with enhanced access control and transfer options.

Dubbing

• API Schema Update: Updated our API documentation to explicitly require the target_language parameter for dubbing projects. This parameter has always been required - we're just making it clearer in our docs. No code changes needed. …

Text to Voice

• Voice Design: Launched new Text to Voice Design with Eleven v3 for creating custom voices from text descriptions.

Speech to Text

• Enhanced Diarization: Added diarization_threshold parameter to the Speech to Text endpoint. Fine-tune the balance between speaker accuracy and total speaker count by adjusting the threshold between 0.1 and 0.4.

Professional Voice Cloning

• Background Noise Removal: Added remove_background_noise to clean up voice samples using audio isolation models for better quality training data.

ElevenCreative Studio

• Video Support Detection: Added has_video property to chapter responses to indicate whether chapters contain video content.

Workspaces

• Service Account Groups: Service accounts can now be added to workspace groups for better permission management and access control.

• Workspace Authentication: Added support for workspace authentication connections, enabling secure webhook tool integrations with external services.

SDKs …

Tools migration

• Agents Platform tools migration: The way tools in Agents Platform are handled is being migrated, please see the guide here to understand what's changing and how to migrate

Text to Speech

• Audio tags automatic removal: Audio tags are now automatically removed when switching from V3 to V2 models, ensuring optimal compatibility and performance.

Agents Platform

• Tools management UI: Added a new comprehensive tools management interface for creating, configuring, and managing tools across all agents in your workspace.
• Streamlined agent creation: Introduced a new agent creation flow with improved user experience and better configuration options.
• Agent duplication: Added the ability to duplicate existing agents, allowing you to quickly create variations of successful agent configurations.

SIP Trunking

• Inbound media encryption: Added support for configurable inbound media encryption settings for SIP trunk phone numbers, enhancing security options.

Voices

• Famous voice category: Added a new "famous" voice category to the voice library, expanding the available voice options for users.

Dubbing …

Agents Platform

• Dynamic variables in simulated conversations: Added support for dynamic variable population in simulated conversations, enabling more flexible and context-aware conversation testing scenarios.
• MCP server integration: Introduced comprehensive support for Model Context Protocol (MCP) servers, allowing agents to connect to external tools and services through standardized protocols with configurable approval policies.
• Burst pricing for extra concurrency: Added bursting capability for workspace call limits, automatically allowing up to 3x the configured concurrency limit during peak usage for overflow capacity.

ElevenCreative Studio

• JSON content initialization: Added support for initializing ElevenCreative Studio projects with structured JSON content through the from_content_json parameter, enabling programmatic project creation with predefined chapters, blocks, and voice configurations.

Workspaces

• Webhook management: Introduced workspace-level webhook management capabilities, allowing administrators to view, configure, and monitor webhook integrations across the entire workspace with detailed usage tracking and failure diagnostics.

API

Text to Speech

• Eleven v3 (alpha): Released Eleven v3 (alpha), our most expressive Text to Speech model, as a research preview.

Agents Platform

• Custom voice settings in multi-voice: Added support for configuring individual voice settings per supported voice in multi-voice agents, allowing fine-tuned control over stability, speed, similarity boost, and streaming latency for each voice.
• Silent transfer to human in Twilio: Added backend configuration support for silent (cold) transfer to number in the Twilio native integration, enabling seamless handoff without announcing the transfer to callers.
• Batch calling retry and cancel: Added support for retrying outbound calls to phone numbers that did not respond during a batch call, along with the ability to cancel ongoing batch operations for better campaign management.
• LLM pinning: Added support for versioned LLM models with explicit checkpoint identifiers
• Custom LLM headers: Added support for passing custom headers to custom LLMs …

Agents Platform

• Multi-voice support for agents: Enable ElevenLabs agents to dynamically switch between different voices during conversations for multi-character storytelling, language tutoring, and role-playing scenarios.
• Claude Sonnet 4 support: Added Claude Sonnet 4 as a new LLM option for conversational agents, providing enhanced reasoning capabilities and improved performance.
• Genesys Cloud integration: Introduced AudioHook Protocol integration for seamless connection with Genesys Cloud contact center platform.
• Force delete knowledge base documents: Added force parameter to knowledge base document deletion, allowing removal of documents even when used by agents.
• Multimodal widget: Added text input and text-only mode defaults for better user experience with improved widget configuration.

API

Forced Aligment

• Forced alignment improvements: Fixed a rare failure case in forced alignment processing to improve reliability.

Voices

• Live moderated voices filter: Added include_live_moderated query parameter to the shared voices endpoint, allowing you to include or exclude voices that are live moderated.

Agents Platform

• Secret dynamic variables: Added support for specifying dynamic variables as secrets with the secret__ prefix. Secret dynamic variables can only be used in webhook tool headers and are never sent to an LLM, enhancing security for sensitive data. Learn more.
• Skip turn system tool: Introduced a new system tool called skip_turn. When enabled, the agent will skip its turn if the user explicitly indicates they need a moment to think or perform an action (e.g., "just a sec", "give me a minute"). This prevents turn timeout from being triggered during intentional user pauses. See the skip turn tool docs for more information.
• Text input support: Added text input support in websocket connections via "user_message" event with text field. Also added "user_activity" event support to indicate typing or other UI activity, improving agent turn-taking when there's interleaved text and audio input. …

SDKs

• SDKs V2: Released new v2 SDKs for both Python and JavaScript

Speech to Text

• Speech to text logprobs: The Speech to Text response now includes a logprob field for word prediction confidence.

Billing

• Improved API error messages: Enhanced API error messages for subscriptions with failed payments. This provides clearer information if a failed payment has caused a user to reach their quota threshold sooner than expected.

Agents Platform

• Batch calls: Released new batch calling functionality, which allows you to automate groups of outbound calls.
• Increased evaluation criteria limit: The maximum number of evaluation criteria for agent performance evaluation has been increased from 5 to 10.
• Human-readable IDs: Introduced human-readable IDs for key Agents Platform entities (e.g., agents, conversations). This improves usability and makes resources easier to identify and manage through the API and UI.
• Unanswered call tracking: 'Not Answered' outbound calls are now reliably detected and visible in the conversation history.
• LLM cost visibility in dashboard: The Agents Platform dashboard now displays the total and per-minute average LLM costs.
• Zero retention mode (ZRM) for agents: Allowed enabling Zero Retention Mode (ZRM) per agent. …

Billing

• Downgraded Plan Pricing Fix: Fixed an issue where customers with downgraded subscriptions were shown their current price instead of the correct future price.

Agents Platform

• Edit Knowledge Base Document Names: You can now edit the names of knowledge base documents. See: Knowledge Base
• Conversation Simulation: Released a new endpoint that allows you to test an agent over text

ElevenCreative Studio

• Export Paragraphs as Zip: Added support for exporting separated paragraphs in a zip file. See: ElevenCreative Studio

SDKs

• Released new SDKs:
  • ElevenLabs Python v1.58.1
  • ElevenLabs JS v1.58.0

API

Dubbing

• Disable Voice Cloning: Added an option in the Dubbing Studio UI to disable voice cloning when uploading audio, aligning with the existing disable_voice_cloning API parameter.

Billing

• Quota Exceeded Error: Improved error messaging for exceeding character limits. Users attempting to generate audio beyond their quota within a short billing window will now receive a clearer 401 unauthorized: This request exceeds your quota limit of... error message indicating the limit has been exceeded.

SDKs

• Released new SDKs: Added ElevenLabs Python v1.58.0 and ElevenLabs JS v1.58.0 to fix a breaking change that had been mistakenly shipped

Agents Platform

• Custom Dashboard Charts: The Agents Platform dashboard can now be extended with custom charts displaying the results of evaluation criteria over time. See the new GET and PATCH endpoints for managing dashboard settings.
• Call History Filtering: Added the ability to filter the call history by start date using the new call_start_before_unix parameter in the List Conversations endpoint. Try it here.
• Server Tools: Added option of making PUT requests in server tools
• Transfer to number: Added call forwarding functionality to support forwarding to operators, see docs here
• Language detection: Fixed an issue where the language detection system tool would trigger on a user replying yes in non-English language.

Usage Analytics

• Custom Aggregation: Added an optional aggregation_interval parameter to the Get Usage Metrics endpoint to control the interval over which to aggregate character usage (hour, day, week, month, or cumulative). …

Professional Voice Cloning (PVC)

• PVC API: Introduced a comprehensive suite of API endpoints for managing Professional Voice Clones (PVC). You can now programmatically create voices, add/manage/delete audio samples, retrieve audio/waveforms, manage speaker separation, handle verification, and initiate training. For a full list of new endpoints check the API changes summary below or read the PVC API reference here.

Speech to Text

• Enhanced Export Options: Added options to include or exclude timestamps and speaker IDs when exporting Speech to Text results in segmented JSON format via the API.

Agents Platform

• New LLM Models: Added support for new GPT-4.1 models: gpt-4.1, gpt-4.1-mini, and gpt-4.1-nanohere
• VAD Score: Added a new client event which sends VAD scores to the client, see reference here

API

Voices

• New PVC flow: Added new flow for Professional Voice Clone creation, try it out here

Agents Platform

• Agent-agent transfer: Added support for agent-to-agent transfers via a new system tool, enabling more complex conversational flows. See the Agent Transfer tool documentation for details.
• Enhanced tool debugging: Improved how tool execution details are displayed in the conversation history for easier debugging.
• Language detection fix: Resolved an issue regarding the forced calling of the language detection tool.

Dubbing

• Render endpoint: Introduced a new endpoint to regenerate audio or video renders for specific languages within a dubbing project. This automatically handles missing transcriptions or translations. See the Render Dub endpoint.
• Increased size limit: Raised the maximum allowed file size for dubbing projects to 1 GiB.

API

Speech to text

• scribe_v1_experimental: Launched a new experimental preview of the Scribe v1 model with improvements including improved performance on audio files with multiple languages, reduced hallucinations when audio is interleaved with silence, and improved audio tags. The new model is available via the API under the model name scribe_v1_experimental

Text to speech

• A-law format support: Added a-law format with 8kHz sample rate to enable integration with European telephony systems.
• Fixed quota issues: Fixed a database bug that caused some requests to be mistakenly rejected as exceeding their quota.

Agents Platform

• Document type filtering: Added support for filtering knowledge base documents by their type (file, URL, or text).
• Non-audio agents: Added support for conversational agents that don't output audio but still send response transcripts and can use tools. Non-audio agents can be enabled by removing the audio client event. …