requests.get() and you have structured earnings data: transcripts, speaker segments and search results as JSON that drops straight into pandas. No SDK to install, no scraping to maintain.
requests, httpx or aiohttp, no SDK required
import requests
API = "https://earningscalls.dev/api/v1"
headers = {"X-API-Key": "ect_your_key_here"}
# Search across all transcripts
hits = requests.get(f"{API}/search/",
params={"q": '"guidance raised"', "limit": 10},
headers=headers).json()
# Full transcript with speaker segments
call_id = hits["data"][0]["earnings_call_id"]
transcript = requests.get(f"{API}/transcripts/{call_id}",
headers=headers).json()
# Into pandas for analysis
import pandas as pd
segments = pd.DataFrame(transcript["data"]["segments"])
print(segments.groupby("speaker_type").size())Four primitives that turn raw earnings calls into a clean dataset your strategy can consume.
Every segment pre-tagged. Five categories with verified counts: executive (826k), analyst (470k), operator (208k), attendee (33k), shareholder (4k). 93% carry the speaker's name.
GET /api/v1/speakers/:idSweep the live archive for phrases like "margin pressure" in one query. Boolean operators, exact phrases, ticker filters across global markets.
GET /api/v1/search?q=…Scan up to 500 tickers in a single request. Per-ticker counts, sector breakdown, last-match date — screen a global universe in one call.
GET /api/v1/search/by_tickerIncremental sync via since / after_id. Re-runnable, idempotent. New calls land within hours of the wire and flow straight into your pipeline.
GET /api/v1/transcripts/recentChange the ticker, click Run, see the actual response.
GET /api/v1/search?q=agentic AI&type=speakers&speaker_type=Executive&ticker=NVDA&limit=5
Read for free. Automate when you're ready. Cancel anytime.
Full access for builders and traders
For production apps and teams
High-volume production workloads
Coverage, query semantics, and what the data supports.
You don't need one - the API is plain REST with predictable JSON, so requests or httpx is all it takes. The docs include copy-paste Python examples for every endpoint group.
Poll /transcripts/recent with cursor pagination - it returns transcripts by ingest time, so a cron job picks up exactly what's new. Ideal for vector stores and research databases.
Yes - responses are flat arrays of records. pd.DataFrame(response["data"]) is usually the whole ETL. Speaker segments, search hits and calendar rows all follow that shape.
Pro: 20 req/min and 5,000/month. Ultra: 60/min, 25,000/month. Enterprise: 120/min, 100,000/month. Details on the earnings call API overview.
What research professionals say about the data and the API.
“Excellent service. Comprehensive coverage of the sectors I wanted and the API is very easy to work with. Highly recommend.”
“Transcripts are accurate, and the API is very user friendly.”
“Connecting Claude through MCP to analyze earnings transcripts has never been easier.”
From the team that brought you the Earnings Calls API - same clean data, same fair pricing.
Pre-configured, industry-standard stock screeners - near real-time, refreshed every 5 minutes, no setup required.
Scans options activity across all US stocks - immediate email alerts the moment institutional-sized trades hit your watchlist.