> ## Documentation Index > Fetch the complete documentation index at: https://docs.mrassistant.ai/llms.txt > Use this file to discover all available pages before exploring further. # Get Transcript > Get full transcript for a specific session. Returns the complete conversation history with all messages, metadata, and AI-generated summary/analysis. The `ai_summary` field contains: - **summary**: Concise overview of the conversation - **sentiment**: Sentiment analysis with score (-1.0 to 1.0) - **topics**: Main topics discussed - **action_items**: Any follow-up actions identified - **key_moments**: Important turning points - **outcome**: How the call concluded (resolved, escalated, etc.) - **call_type**: Type of call (support, sales, inquiry, etc.) Set `include_ai_summary=false` to skip AI analysis and reduce latency. ## OpenAPI ````yaml https://api.mrassistant.ai/openapi.json get /agents/{agent_id}/transcripts/{session_id} openapi: 3.1.0 info: title: Backend - Multilingual Voice Agent System description: FastAPI backend for multilingual voice agents with tool use version: 1.0.0 servers: [] security: [] paths: /agents/{agent_id}/transcripts/{session_id}: get: tags: - agent-transcripts summary: Get Transcript description: |- Get full transcript for a specific session. Returns the complete conversation history with all messages, metadata, and AI-generated summary/analysis. The `ai_summary` field contains: - **summary**: Concise overview of the conversation - **sentiment**: Sentiment analysis with score (-1.0 to 1.0) - **topics**: Main topics discussed - **action_items**: Any follow-up actions identified - **key_moments**: Important turning points - **outcome**: How the call concluded (resolved, escalated, etc.) - **call_type**: Type of call (support, sales, inquiry, etc.) Set `include_ai_summary=false` to skip AI analysis and reduce latency. operationId: get_transcript_agents__agent_id__transcripts__session_id__get parameters: - name: agent_id in: path required: true schema: type: string format: uuid title: Agent Id - name: session_id in: path required: true schema: type: string format: uuid title: Session Id - name: include_ai_summary in: query required: false schema: type: boolean description: Generate AI summary using OpenAI (adds ~1-2s latency) default: true title: Include Ai Summary description: Generate AI summary using OpenAI (adds ~1-2s latency) responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/TranscriptResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - HTTPBearer: [] components: schemas: TranscriptResponse: properties: session_id: type: string title: Session Id agent_id: type: string title: Agent Id agent_name: type: string title: Agent Name channel: type: string title: Channel phone_number: anyOf: - type: string - type: 'null' title: Phone Number caller_number: anyOf: - type: string - type: 'null' title: Caller Number start_time: type: string title: Start Time end_time: anyOf: - type: string - type: 'null' title: End Time duration_seconds: anyOf: - type: integer - type: 'null' title: Duration Seconds status: type: string title: Status messages: items: $ref: '#/components/schemas/TranscriptMessage' type: array title: Messages summary: anyOf: - type: string - type: 'null' title: Summary sentiment_score: anyOf: - type: number - type: 'null' title: Sentiment Score language_detected: anyOf: - type: string - type: 'null' title: Language Detected ai_summary: anyOf: - $ref: '#/components/schemas/AITranscriptSummary' - type: 'null' description: >- AI-generated summary and analysis of the conversation (generated on demand) type: object required: - session_id - agent_id - agent_name - channel - start_time - status - messages title: TranscriptResponse description: Full transcript response with AI summary. HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError TranscriptMessage: properties: timestamp: type: string title: Timestamp role: type: string title: Role text: type: string title: Text lang: type: string title: Lang mt_from_lang: anyOf: - type: string - type: 'null' title: Mt From Lang mt_to_lang: anyOf: - type: string - type: 'null' title: Mt To Lang payload: anyOf: - additionalProperties: true type: object - type: 'null' title: Payload type: object required: - timestamp - role - text - lang title: TranscriptMessage description: >- A single message in the transcript. The 'role' field differentiates between speakers: - "user" or "human": Customer/caller speaking - "assistant" or "agent": AI agent speaking Frontend should use 'role' to style messages differently (e.g., alignment, colors). See TRANSCRIPT_UI_GUIDE.md for UI implementation examples. example: lang: en role: user text: I need help with my account timestamp: '2025-12-05T10:00:12Z' AITranscriptSummary: properties: summary: type: string title: Summary description: Concise summary of the conversation sentiment: anyOf: - $ref: '#/components/schemas/SentimentAnalysis' - type: 'null' description: Sentiment analysis topics: items: type: string type: array title: Topics description: Main topics discussed action_items: items: type: string type: array title: Action Items description: Follow-up actions identified key_moments: items: type: string type: array title: Key Moments description: Key moments in the conversation outcome: anyOf: - type: string - type: 'null' title: Outcome description: 'Call outcome: resolved, escalated, etc.' call_type: anyOf: - type: string - type: 'null' title: Call Type description: 'Type of call: support, sales, inquiry, etc.' type: object required: - summary title: AITranscriptSummary description: AI-generated transcript summary and analysis. example: action_items: - Send confirmation email call_type: support key_moments: - Agent offered a 10% credit which resolved the issue outcome: resolved sentiment: agent_performance: Agent handled the situation professionally customer_satisfaction: Customer was satisfied with the resolution overall: positive score: 0.7 summary: >- Customer called about a billing issue. Agent resolved it by applying a credit. topics: - billing - account credit ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError SentimentAnalysis: properties: overall: type: string title: Overall description: 'Overall sentiment: positive, neutral, negative, mixed' score: type: number title: Score description: Sentiment score from -1.0 (negative) to 1.0 (positive) customer_satisfaction: anyOf: - type: string - type: 'null' title: Customer Satisfaction description: Assessment of customer satisfaction agent_performance: anyOf: - type: string - type: 'null' title: Agent Performance description: Assessment of agent handling type: object required: - overall - score title: SentimentAnalysis description: AI-generated sentiment analysis. securitySchemes: HTTPBearer: type: http scheme: bearer ````