@elevenlabs/elevenlabs-js 2.0.1 → 2.1.0

package/.mock/asyncapi.yml

@@ -147,7 +147,7 @@ channels:
       contexts, flush them, or close them independently. A `close_socket` message can be used to terminate
       the entire connection gracefully.
 
-      For more information on how to use this API for conversational agents see the [conversational agents guide](/docs/best-practices/conversational-agents).
+      For more information on best practices for how to use this API, please see the [multi context websocket guide](/docs/cookbooks/multi-context-web-socket).
 
     bindings:
       ws:
@@ -624,18 +624,18 @@ components:
           items:
             $ref: '#/components/schemas/PronunciationDictionaryLocator'
           description: Optional list of pronunciation dictionary locators. Can only be provided in the first message for a given context_id.
-        context_id:
+        contextId:
           type: string
           nullable: true
           description: An identifier for the text-to-speech context. Allows managing multiple independent audio generation streams over a single WebSocket connection. If omitted, a default context is used.
         close_context:
           type: boolean
           default: false
-          description: If true, closes the specified `context_id`. No further audio will be generated for this context. The `text` field is ignored.
+          description: If true, closes the specified `contextId`. No further audio will be generated for this context. The `text` field is ignored.
         close_socket:
           type: boolean
           default: false
-          description: If true, flushes all contexts and closes the entire WebSocket connection. The `text` and `context_id` fields are ignored.
+          description: If true, flushes all contexts and closes the entire WebSocket connection. The `text` and `contextId` fields are ignored.
 
     WebsocketTTSServerMessageMulti:
       type: object
@@ -762,9 +762,9 @@ components:
         alignment:
           $ref: '#/components/schemas/Alignment'
           nullable: true
-        context_id:
+        contextId:
           type: string
-          description: The context_id for which this audio is.
+          description: The contextId for which this audio is.
 
     FinalOutputMulti:
       type: object

package/.mock/definition/__package__.yml

@@ -1233,8 +1233,19 @@ types:
     properties: {}
     source:
       openapi: openapi.json
+  ChapterContentBlockInputModelSubType:
+    enum:
+      - p
+      - h1
+      - h2
+      - h3
+    inline: true
+    source:
+      openapi: openapi.json
   ChapterContentBlockInputModel:
     properties:
+      sub_type:
+        type: optional<ChapterContentBlockInputModelSubType>
       block_id:
         type: optional<string>
       nodes:
@@ -2162,6 +2173,9 @@ types:
       expiration_time_unix_secs:
         type: optional<integer>
         docs: The expiration time of the token in unix seconds
+      conversation_id:
+        type: optional<string>
+        docs: The ID of the conversation
       purpose:
         type: optional<ConversationTokenPurpose>
         docs: The purpose of the token
@@ -2997,6 +3011,11 @@ types:
         type: WidgetConfigResponseModel
     source:
       openapi: openapi.json
+  GetAgentKnowledgebaseSizeResponseModel:
+    properties:
+      number_of_pages: double
+    source:
+      openapi: openapi.json
   GetAgentLinkResponseModel:
     properties:
       agent_id:
@@ -3631,6 +3650,19 @@ types:
         type: optional<map<string, LlmInputOutputTokensUsage>>
     source:
       openapi: openapi.json
+  LlmUsageCalculatorLlmResponseModel:
+    properties:
+      llm:
+        type: Llm
+      price_per_minute: double
+    source:
+      openapi: openapi.json
+  LlmUsageCalculatorResponseModel:
+    properties:
+      llm_prices:
+        type: list<LlmUsageCalculatorLlmResponseModel>
+    source:
+      openapi: openapi.json
   LanguageAddedResponse:
     properties:
       version: integer
@@ -4047,6 +4079,66 @@ types:
         docs: Whether the user is on the watchlist.
     source:
       openapi: openapi.json
+  NativeMcpToolConfigInput:
+    docs: A Native MCP tool is a tool that is used to call a Native MCP server
+    properties:
+      id:
+        type: optional<string>
+        default: ''
+      name:
+        type: string
+        validation:
+          pattern: ^[a-zA-Z0-9_-]{1,64}$
+          minLength: 0
+      description:
+        type: string
+        validation:
+          minLength: 0
+      response_timeout_secs:
+        type: optional<integer>
+        docs: The maximum time in seconds to wait for the tool call to complete.
+        default: 20
+      parameters:
+        type: optional<ObjectJsonSchemaPropertyInput>
+        docs: Schema for any parameters the LLM needs to provide to the MCP tool.
+      mcp_tool_name:
+        type: string
+        docs: The name of the MCP tool to call
+      mcp_server_id:
+        type: string
+        docs: The id of the MCP server to call
+    source:
+      openapi: openapi.json
+  NativeMcpToolConfigOutput:
+    docs: A Native MCP tool is a tool that is used to call a Native MCP server
+    properties:
+      id:
+        type: optional<string>
+        default: ''
+      name:
+        type: string
+        validation:
+          pattern: ^[a-zA-Z0-9_-]{1,64}$
+          minLength: 0
+      description:
+        type: string
+        validation:
+          minLength: 0
+      response_timeout_secs:
+        type: optional<integer>
+        docs: The maximum time in seconds to wait for the tool call to complete.
+        default: 20
+      parameters:
+        type: optional<ObjectJsonSchemaPropertyOutput>
+        docs: Schema for any parameters the LLM needs to provide to the MCP tool.
+      mcp_tool_name:
+        type: string
+        docs: The name of the MCP tool to call
+      mcp_server_id:
+        type: string
+        docs: The id of the MCP server to call
+    source:
+      openapi: openapi.json
   ObjectJsonSchemaPropertyInputPropertiesValue:
     discriminated: false
     union:
@@ -4653,6 +4745,8 @@ types:
         type: ClientToolConfigInput
       mcp:
         type: McpToolConfigInput
+      native_mcp:
+        type: NativeMcpToolConfigInput
       system:
         type: SystemToolConfigInput
       webhook:
@@ -4687,6 +4781,9 @@ types:
       mcp_server_ids:
         type: optional<list<string>>
         docs: A list of MCP server ids to be used by the agent
+      native_mcp_server_ids:
+        type: optional<list<string>>
+        docs: A list of Native MCP server ids to be used by the agent
       knowledge_base:
         type: optional<list<KnowledgeBaseLocator>>
         docs: A list of knowledge bases to be used by the agent
@@ -4710,6 +4807,8 @@ types:
         type: ClientToolConfigOutput
       mcp:
         type: McpToolConfigOutput
+      native_mcp:
+        type: NativeMcpToolConfigOutput
       system:
         type: SystemToolConfigOutput
       webhook:
@@ -4725,6 +4824,8 @@ types:
         type: ClientToolConfigInput
       mcp:
         type: McpToolConfigInput
+      native_mcp:
+        type: NativeMcpToolConfigInput
       system:
         type: SystemToolConfigInput
       webhook:
@@ -4759,6 +4860,9 @@ types:
       mcp_server_ids:
         type: optional<list<string>>
         docs: A list of MCP server ids to be used by the agent
+      native_mcp_server_ids:
+        type: optional<list<string>>
+        docs: A list of Native MCP server ids to be used by the agent
       knowledge_base:
         type: optional<list<KnowledgeBaseLocator>>
         docs: A list of knowledge bases to be used by the agent
@@ -7486,7 +7590,7 @@ types:
         docs: >-
           Optional list of pronunciation dictionary locators. Can only be
           provided in the first message for a given context_id.
-      context_id:
+      contextId:
         type: optional<string>
         docs: >-
           An identifier for the text-to-speech context. Allows managing multiple
@@ -7495,14 +7599,14 @@ types:
       close_context:
         type: optional<boolean>
         docs: >-
-          If true, closes the specified `context_id`. No further audio will be
+          If true, closes the specified `contextId`. No further audio will be
           generated for this context. The `text` field is ignored.
         default: false
       close_socket:
         type: optional<boolean>
         docs: >-
           If true, flushes all contexts and closes the entire WebSocket
-          connection. The `text` and `context_id` fields are ignored.
+          connection. The `text` and `contextId` fields are ignored.
         default: false
     source:
       openapi: asyncapi.yml
@@ -7631,9 +7735,9 @@ types:
         docs: Base64 encoded audio chunk.
       normalizedAlignment: optional<NormalizedAlignment>
       alignment: optional<Alignment>
-      context_id:
+      contextId:
         type: optional<string>
-        docs: The context_id for which this audio is.
+        docs: The contextId for which this audio is.
     source:
       openapi: asyncapi.yml
   FinalOutputMulti:

package/.mock/definition/conversationalAi/agents/knowledgeBase.yml

@@ -0,0 +1,32 @@
+imports:
+  root: ../../__package__.yml
+service:
+  auth: false
+  base-path: ''
+  endpoints:
+    size:
+      path: /v1/convai/agent/{agent_id}/knowledge-base/size
+      method: GET
+      auth: false
+      docs: Returns the number of pages in the agent's knowledge base.
+      source:
+        openapi: openapi.json
+      path-parameters:
+        agent_id: string
+      display-name: Returns The Size Of The Agent'S Knowledge Base
+      response:
+        docs: Successful Response
+        type: root.GetAgentKnowledgebaseSizeResponseModel
+        status-code: 200
+      errors:
+        - root.UnprocessableEntityError
+      examples:
+        - path-parameters:
+            agent_id: agent_id
+          response:
+            body:
+              number_of_pages: 1.1
+      audiences:
+        - convai
+  source:
+    openapi: openapi.json

package/.mock/definition/conversationalAi/agents/link.yml

@@ -29,9 +29,10 @@ service:
             body:
               agent_id: J3Pbu5gP6NNKBscdCdwB
               token:
-                agent_id: J3Pbu5gP6NNKBscdCdwB
+                agent_id: agent_J3Pbu5gP6NNKBscdCdwB
                 conversation_token: '1234567890'
                 expiration_time_unix_secs: 1716153600
+                conversation_id: conv_J3Pbu5gP6NNKBscdCdwB
                 purpose: signed_url
       audiences:
         - convai

package/.mock/definition/conversationalAi/agents/llmUsage.yml

@@ -0,0 +1,51 @@
+imports:
+  root: ../../__package__.yml
+service:
+  auth: false
+  base-path: ''
+  endpoints:
+    calculate:
+      path: /v1/convai/agent/{agent_id}/llm-usage/calculate
+      method: POST
+      auth: false
+      docs: Calculates expected number of LLM tokens needed for the specified agent.
+      source:
+        openapi: openapi.json
+      path-parameters:
+        agent_id: string
+      display-name: Calculate Expected Llm Usage For An Agent
+      request:
+        name: LlmUsageCalculatorRequestModel
+        body:
+          properties:
+            prompt_length:
+              type: optional<integer>
+              docs: Length of the prompt in characters.
+            number_of_pages:
+              type: optional<integer>
+              docs: >-
+                Pages of content in pdf documents OR urls in agent's Knowledge
+                Base.
+            rag_enabled:
+              type: optional<boolean>
+              docs: Whether RAG is enabled.
+        content-type: application/json
+      response:
+        docs: Successful Response
+        type: root.LlmUsageCalculatorResponseModel
+        status-code: 200
+      errors:
+        - root.UnprocessableEntityError
+      examples:
+        - path-parameters:
+            agent_id: agent_id
+          request: {}
+          response:
+            body:
+              llm_prices:
+                - llm: gpt-4o-mini
+                  price_per_minute: 1.1
+      audiences:
+        - convai
+  source:
+    openapi: openapi.json

package/.mock/definition/conversationalAi/llmUsage.yml

@@ -0,0 +1,52 @@
+imports:
+  root: ../__package__.yml
+service:
+  auth: false
+  base-path: ''
+  endpoints:
+    calculate:
+      path: /v1/convai/llm-usage/calculate
+      method: POST
+      auth: false
+      docs: >-
+        Returns a list of LLM models and the expected cost for using them based
+        on the provided values.
+      source:
+        openapi: openapi.json
+      display-name: Calculate Expected Llm Usage
+      request:
+        name: LlmUsageCalculatorPublicRequestModel
+        body:
+          properties:
+            prompt_length:
+              type: integer
+              docs: Length of the prompt in characters.
+            number_of_pages:
+              type: integer
+              docs: >-
+                Pages of content in PDF documents or URLs in the agent's
+                knowledge base.
+            rag_enabled:
+              type: boolean
+              docs: Whether RAG is enabled.
+        content-type: application/json
+      response:
+        docs: Successful Response
+        type: root.LlmUsageCalculatorResponseModel
+        status-code: 200
+      errors:
+        - root.UnprocessableEntityError
+      examples:
+        - request:
+            prompt_length: 1
+            number_of_pages: 1
+            rag_enabled: true
+          response:
+            body:
+              llm_prices:
+                - llm: gpt-4o-mini
+                  price_per_minute: 1.1
+      audiences:
+        - convai
+  source:
+    openapi: openapi.json

package/.mock/definition/v1TextToSpeechVoiceIdMultiStreamInput.yml

@@ -26,9 +26,9 @@ channel:
     the entire connection gracefully.
 
 
-    For more information on how to use this API for conversational agents see
-    the [conversational agents
-    guide](/docs/best-practices/conversational-agents).
+    For more information on best practices for how to use this API, please see
+    the [multi context websocket
+    guide](/docs/cookbooks/multi-context-web-socket).
   path-parameters:
     voice_id:
       type: string
@@ -192,6 +192,7 @@ channel:
                 - r
                 - l
                 - d
+            contextId: conv_1
         - type: publish
           body:
             text: 'Hi this is a new context with different settings! '
@@ -210,6 +211,7 @@ channel:
         - type: subscribe
           body:
             audio: Y3VyaW91cyBtaW5kcyB0aGluayBhbGlrZSA6KQ==
+            contextId: interruption_context
         - type: subscribe
           body:
             isFinal: true