> ## Documentation Index > Fetch the complete documentation index at: https://docs.goparlay.io/llms.txt > Use this file to discover all available pages before exploring further. # Submit a long-form recording for boundary detection + per-segment analysis > Upload one HTTPS URL of a multi-hour recording. The server detects conversation boundaries (single-shot Gemini), then auto-analyzes each high-confidence segment with the V5 master prompt. Each segment becomes a first-class Analysis row with `parent_session_id` set, queryable via the existing `/v1/analyses` surface. **Audio storage:** we never re-host the file. The response contains `start_seconds`/`end_seconds` offsets — your UI uses HTTP Range requests on your original URL to play any segment. See `/guides/all-day-sessions` for the full pattern. **Pricing:** ~$0.30 per hour of audio for detection (Deepgram + Gemini boundary call) plus ~$0.30 per detected segment for V5 analysis. ## OpenAPI ````yaml https://api.goparlay.io/docs/json post /v1/all-day-sessions openapi: 3.0.3 info: title: Parlay API description: >- Partner-facing API for Parlay pitch scoring, rep intelligence, and coaching. See API.md for the full spec. version: 0.1.0 servers: - url: http://localhost:3000 description: production security: - ApiKeyAuth: [] tags: - name: Health description: Service health and platform status - name: Analyses description: Submit and retrieve pitch scoring analyses - name: Orgs description: Partner org management - name: Reps description: Rep management - name: Keys description: API key lifecycle - name: Webhooks description: Webhook registration and delivery - name: Personas description: Rep persona intelligence - name: Methodologies description: Rep methodology intelligence - name: Synthesis description: Profile synthesis - name: Insights description: Org-level insights - name: Playbooks description: Playbook management + AI drafting - name: Prompts description: Custom AI prompts - name: Scoring description: Scoring profile config - name: Chat description: Scout AI chat - name: Dispositions description: Call dispositions - name: Practice description: Practice sessions - name: Bulk description: Batch and export - name: Usage description: Usage and billing - name: Audit description: Audit log paths: /v1/all-day-sessions: post: tags: - All-day sessions summary: >- Submit a long-form recording for boundary detection + per-segment analysis description: >- Upload one HTTPS URL of a multi-hour recording. The server detects conversation boundaries (single-shot Gemini), then auto-analyzes each high-confidence segment with the V5 master prompt. Each segment becomes a first-class Analysis row with `parent_session_id` set, queryable via the existing `/v1/analyses` surface. **Audio storage:** we never re-host the file. The response contains `start_seconds`/`end_seconds` offsets — your UI uses HTTP Range requests on your original URL to play any segment. See `/guides/all-day-sessions` for the full pattern. **Pricing:** ~$0.30 per hour of audio for detection (Deepgram + Gemini boundary call) plus ~$0.30 per detected segment for V5 analysis. requestBody: required: true content: application/json: schema: type: object properties: recording_url: type: string format: uri description: >- Publicly reachable HTTPS URL of the long-form audio. Pre-signed S3/GCS URLs are perfect. Audio stays on partner storage — we never re-host the file. Mutually exclusive with file_key. file_key: type: string minLength: 1 maxLength: 512 description: >- GCS key returned from POST /v1/admin/partners/:id/uploads/signed-url after a browser-direct PUT. The server resolves this to a 1-hour signed read URL internally. Mutually exclusive with recording_url. org_id: type: string minLength: 1 maxLength: 256 pattern: ^[A-Za-z0-9._:-]+$ description: Partner-supplied org id this session belongs to. rep_id: type: string minLength: 1 maxLength: 256 pattern: ^[A-Za-z0-9._:-]+$ description: >- Partner-supplied rep id. Required — every segment analysis is scoped to this rep. options: type: object properties: auto_analyze: type: boolean description: >- Default true. When true, every detected segment with confidence >= min_confidence is auto-submitted for V5 analysis. When false, segments are detected only — partner triggers each analysis manually via POST /v1/all-day-sessions/:id/segments/:seg_id/analyze. min_confidence: type: number minimum: 0 maximum: 1 description: >- Default 0.8. Segments at or above this confidence auto-analyze. Segments below 0.5 are skipped entirely. Segments in between go to status='needs_review' and the partner can decide whether to analyze them. callback_url: type: string format: uri description: >- Optional ad-hoc webhook URL for THIS session only. Receives all_day_session.detected/completed/failed events. Use registered webhooks (POST /v1/webhooks) for the standard partner-wide subscription. language: type: string description: ISO language code for Deepgram. Auto-detect if omitted. skip_diarization: type: boolean description: >- Default false. Diarization is required for boundary detection — only set to true if you know the audio is single-speaker monologue. playbook_id: type: string format: uuid description: >- Override the active playbook for every segment in this session. prompt_overrides: type: array items: type: object properties: title: type: string content: type: string required: - title - content additionalProperties: false description: >- Inline custom prompts applied to every segment in this session. additionalProperties: true metadata: type: object additionalProperties: {} description: >- Free-form key/value bag for partner bookkeeping. Not interpreted by Parlay. required: - org_id - rep_id additionalProperties: false responses: '202': description: Default Response content: application/json: schema: type: object properties: id: type: string format: uuid status: type: string enum: - queued - transcribing - detecting - analyzing - completed - failed created_at: type: string required: - id - status - created_at additionalProperties: false components: securitySchemes: ApiKeyAuth: type: http scheme: bearer bearerFormat: API Key (pk_sandbox_* or pk_live_*) ````