Overview

The FinIQ Transcripts API converts unstructured earnings calls into machine-readable datasets. We provide fully parsed transcripts with Speaker Diarization (identifying who is speaking), Role Mapping (CEO vs. Analyst), and Slide Synchronization.

Key Capabilities:

• History: 15+ Years (US), 10+ Years (Global).
• Structure: JSON-formatted with "Management" and "Q&A" segmentation.
• NLP Ready: Pre-calculated sentiment scores and concept tags.
• Multi-Modal: Links to specific slides and audio timestamps.

‍

1. Access Methods

‍

2. REST API: Transcripts Endpoint

Endpoint:GET https://api.finiq.data/v1/transcripts

Request Parameters:

‍

Example Request (Python)

import requests

url = "https://api.finiq.data/v1/transcripts"
params = {
    "api_token": "YOUR_KEY",
    "ticker": "UBER",
    "year": 2024,
    "quarter": 3
}

response = requests.get(url, params=params)
data = response.json()

‍

Response Structure (JSON)

{
  "ticker": "UBER",
  "quarter": "2024-Q3",
  "date": "2024-11-05",
  "presentation_url": "https://finiq.data/decks/uber_q3_24.pdf",
  "segments": [
    {
      "segment_type": "Management_Remarks",
      "speaker_name": "Dara Khosrowshahi",
      "speaker_role": "CEO",
      "text": "We are seeing unprecedented demand in the mobility segment...",
      "sentiment_score": 0.85,
      "linked_slide": 4
    },
    {
      "segment_type": "Q&A",
      "speaker_name": "Analyst (Goldman Sachs)",
      "text": "Can you elaborate on the margin compression in freight?",
      "sentiment_score": -0.12
    }
  ]
}

‍

3. Data Dictionary (Key Metrics)

‍

4. Bulk Data Access (JSONL)

Ideal for training Large Language Models (LLMs) on financial domain text.

File Naming Convention:

‍{Ticker}_{Year}_Q{Quarter}_Transcript.jsonl

Format (JSON Lines):

‍Each line represents one text segment, allowing for streaming processing of massive datasets without loading the entire file into memory.

How to Download:

1. Navigate to Data Export.
2. Select Transcripts Package.
3. Choose Universe and Date Range.
4. Click Download ZIP.

‍

5. Vector Database Feed (RAG-Ready)

Exclusive to FinIQ.This feed delivers transcripts pre-chunked and embedded, ready for vector search.

Sample Vector Output (Decoded):

"Context: UBER Q3 2024 Q&A. Speaker: CFO. Text: 'Freight remains a cyclical headwind, but we expect EBITDA breakeven by Q4 due to cost rationalization.' Tags: [Guidance, Freight, Cost_Cutting]"

Use Case:

• "Build a chatbot that answers questions based ONLY on what the CFO said."
• "Find all instances of 'AI CapEx' across the Mag 7 transcripts."

‍

6. Error Handling

• 400 Bad Request: Invalid Ticker or Date.
• 401 Unauthorized: Invalid API Key.
• 404 Not Found: Transcript not yet available (e.g., call hasn't happened).
• 429 Too Many Requests: Rate limit exceeded.

‍

7. Need Help?

• Developer Support: email dev-support@finiq.data
• Slack Community: Join our [Quant Developer Slack]

‍