Call
Coaching
API
Build your own call-coaching and review platform on open-source meeting transcription. Transcribe every Meet, Teams, or Zoom call; score it with the AI of your choice; pay compute COGS instead of seat licenses.
Waiting for meeting participants…
Feed meetings into your AI. Live.
AI needs data, and the best data is real-time. Vexa streams transcripts, audio, and context from any meeting straight to your models and agents — as it happens.
Built for open-source call coaching
Vexa is the transcription primitive under call-coaching platforms — the same role Gong / Chorus / Salesloft Conversations play internally. Difference: Vexa is Apache 2.0 and self-hostable, so the audio + transcripts stay yours, and the cost model is compute COGS instead of $1k/seat/year. Most coaching-tool customers we hear from outgrow per-seat pricing somewhere between 20-50 reps; this is what they migrate to. Vexa handles the heavy part (meeting joining, audio capture, Whisper transcription, speaker diarization); you keep ownership of the scorecard / dashboard / workflow layer that's specific to your sales motion.
- Open-source alternative to Gong / Chorus / Salesloft Conversations — Apache 2.0, no per-seat fees
- Transcripts across Google Meet, Microsoft Teams, and Zoom from one API
- Bring your own scoring layer: Einstein, Agentforce, OpenAI, Anthropic, or in-house models
# 1. Trigger Vexa per scheduled call (from CRM or calendar automation)
curl -X POST "$API_BASE/bots" \
-H "X-API-Key: $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"platform": "google_meet",
"native_meeting_id": "abc-defg-hij",
"transcribe_enabled": true,
"transcription_tier": "realtime"
}'
# 2. After the call: fetch the full transcript
curl "$API_BASE/transcripts/google_meet/abc-defg-hij" \
-H "X-API-Key: $API_KEY" > transcript.json
# 3. Score it with your AI of choice (Anthropic shown)
curl https://api.anthropic.com/v1/messages \
-H "x-api-key: $ANTHROPIC_API_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "content-type: application/json" \
-d '{
"model": "claude-opus-4-5",
"max_tokens": 2048,
"messages": [{
"role": "user",
"content": "Score this sales call on MEDDIC. Return JSON: {economic_buyer_identified, metrics_quantified, decision_criteria, decision_process, identify_pain, champion}. Transcript: ..."
}]
}'Set $API_BASE to https://api.cloud.vexa.ai (hosted) or http://localhost:8056 (self-host). Get $API_KEY from /get-started.
Coaching platform FAQ
What sales leaders ask before replacing seat-based coaching SaaS with self-built tooling on Vexa.
Is this really a Gong / Chorus / Salesloft Conversations alternative?
For the transcription + storage layer, yes — that's the heavy part Gong builds and protects. For the scorecard / coaching-dashboard / workflow layer, you assemble it (or use an existing AI tool with Vexa transcripts as input). Honest tradeoff: more glue code to write vs Gong's turnkey UI, vastly different cost curve at scale.
Where does the cost crossover happen?
Most teams we hear from cross over somewhere between 20-50 reps. Gong/Chorus pricing scales with seats (~$1k-1.5k/rep/year for the call-coaching tier). Vexa cost scales with compute (hosted: $0.30-0.45/bot-hour; self-host: compute COGS at maybe $0.05-0.15/hr if you run your own GPUs). Run the math on your call volume.
Can I keep my current call-coaching workflow / scorecards / dashboards?
Yes — that's the point. The scoring + UI + workflow stays whatever you build (or already have); Vexa just replaces the transcription primitive underneath. Common pattern: Vexa transcribes → your scoring service (Einstein, Agentforce, OpenAI, in-house) → your existing dashboard (Salesforce, Looker, Tableau, custom).
What about live coaching during the call (Gong Assist style)?
Subscribe to Vexa's WebSocket stream. Transcript chunks land sub-3-seconds; pipe through an LLM with the rep's playbook in context; surface prompts in your internal app. Sub-5-second end-to-end latency is achievable. You build the live UI; Vexa provides the live stream.
Are there industries where self-host is actually required for coaching?
Yes — regulated industries (financial advice, healthcare consults, legal intake, government) typically can't send call audio to a US-based SaaS at all. Self-hosted Vexa keeps audio and transcripts entirely on infrastructure you control — pair with your existing audit trail and you have a compliant call-coaching pipeline.
Common patterns
What coaching teams build with Vexa
Three patterns sales-leadership and RevOps teams build replacing or supplementing Gong / Chorus.
Self-hosted Gong replacement at 100+ reps
Scenario
You're paying $100k+/year for seat-based call coaching. Sales leadership wants the same workflows but the contract is up for renewal and you have engineering bandwidth.
What you get
Vexa transcribes every call across Meet/Teams/Zoom; your existing scoring logic runs on the transcripts; your existing dashboards consume the structured output. Net cost drops 70-90% at this scale. Audio data stays under your control — a compliance win as well as a cost win.
Real-time AE prompts during discovery calls
Scenario
You want reps to get live prompts on their screen — "ask about decision timeline", "this competitor was mentioned, here's a battlecard" — while the call is happening.
What you get
WebSocket stream of transcript chunks → LLM with rep playbook + battlecards → prompt surfaces in your internal app within 5 seconds. Reps run better discovery without leaving the call to check Notion.
Coaching reviews with timestamped jump-to-moment
Scenario
Managers want to review the top 3 moments per call (objection handled, pricing landed, missed next step) — without scrubbing the full hour-long recording.
What you get
Post-call AI step scores the transcript, picks top moments, attaches timestamps + transcript excerpts to a Coaching Review record. Managers see a 30-second jump-to summary per call. Build the workflow in Notion, Linear, Salesforce, or your in-house tool.
Quickstart
Build your coaching platform in 5 minutes
Vexa is the transcription primitive; your scoring + dashboards stay yours. Wire it up in an afternoon.
- 1
Get a Vexa API key
Sign up at vexa.ai/get-started or run the self-hosted gateway and mint one locally.
- 2
Wire the trigger
Fire Vexa per scheduled meeting from your CRM Workflow, calendar automation, or scheduling tool.
- 3
Capture transcripts
Webhook chunks (real-time) or GET /transcripts (post-call). Choose based on whether you need live coaching or batch review.
- 4
Add the scoring layer
Pipe transcripts to your AI of choice — Einstein, Agentforce, OpenAI, Anthropic, fine-tuned in-house model — with your scorecard prompt or schema.
- 5
Plug into your dashboards
Push structured output to CRM custom fields, BI dashboards, Notion, or your internal coaching tool. Reuse what already works.
How it works under the hood
call coaching architecture
Vexa's role: meeting joining, audio capture, Whisper transcription, speaker diarization, transcript storage. Your role: scoring, dashboards, workflows, alerts. This split is exactly how Gong / Chorus internally separate concerns — the difference is you own both layers and can swap either at will. Self-host moves the audio path entirely inside your VPC for compliance-sensitive coaching workflows.
Pipeline
Documentation
Go deeper on coaching workflows
Transcripts API, WebSocket streaming, scoring patterns, real-time prompt latency.
Transcripts API
Fetch full call transcripts after meetings end — the core input to your scoring layer.
WebSocket usage
Live transcript streaming — for real-time coaching and live prompts.
Webhooks
Push transcript chunks to your endpoint as they happen.
Speaker identification
Rep-vs-customer turn attribution — fundamental for scoring talk ratio.
Transcription quality
Whisper model tiers + accuracy bounds — pick the right tier for your accuracy/cost target.
Self-host deployment
For regulated industries where call audio cannot leave your infrastructure.
Full reference at docs.vexa.ai.
Related reading
Keep going on call coaching
Tutorials, deep-dives, and adjacent surfaces from across the Vexa substrate.
Vexa vs Recall.AI comparison
License, pricing, and self-host differences vs the closed-source incumbent.
Sales call transcription use case
The sibling cluster — feed transcripts into HubSpot/Salesforce deal records.
Open-source transcription API — complete guide
How Vexa's transcription primitive works under any coaching workflow.
Pay only for
what you use
Self-host for free, or let us run it for you. Simple usage-based pricing—no per-seat tax.
Apache 2.0 · Self-host forever
Best for teams who need full control
1 bot · Flat monthly · Everything included
Best for personal use — 1 meeting at a time
No credit card required
Bot infrastructure · +$0.20/hr transcription
Best for teams & API builders · $5 free credit
No credit card required
On-premises, OpenShift, Kubernetes. Dedicated support + SLA.
For self-hosted Vexa bot users. Transcription only — $0.002/min.
Frequently asked questions
What's included in the $0.30/hr Bot Service?
The base rate covers bot infrastructure: audio capture, webhooks, and 12 months of audio storage across Google Meet, Microsoft Teams, and Zoom. Add real-time transcription for +$0.20/hr ($0.50/hr total).
Which plan is right for me?
Choose Individual ($12/mo) if you need one bot for personal use—it includes real-time transcription, storage, and the Dashboard. Choose Pay-as-you-go ($0.30/hr) if you need multiple simultaneous bots or want to pay only for what you use—ideal for teams and API integrations. Both plans include Google Meet and Microsoft Teams support. Self-host for free from GitHub if you need full data control.
How does the free credit work?
Every new account gets $5 in free bot credit—no credit card required. That covers ~16 hours of bot time at $0.30/hr. All features are available: audio capture, transcription, real-time data, and full API access.
Can I self-host Vexa?
Yes. Vexa is Apache 2.0 licensed and fully self-hostable. Deploy with Docker Compose, Kubernetes, or OpenShift on your own infrastructure with complete data sovereignty.
How does Vexa compare to Recall.ai?
Vexa is open source, self-hostable, and up to 40% cheaper. Bot: $0.30/hr vs Recall.ai ~$0.50/hr. Transcription add-on: $0.20/hr vs ~$0.15/hr. See our detailed comparison.
What's the Individual plan for?
The Individual plan ($12/mo) is for single users who need one concurrent bot with unlimited meetings. It includes real-time transcription, recording, storage, REST API, WebSockets, and the UI Dashboard—everything you need for personal meeting intelligence. It supports Google Meet and Microsoft Teams, with Zoom coming soon.
Ready when you are
Coach every rep.
Without the $1k/seat tax.
No credit card on Vexa side, no per-seat coaching-SaaS contract. Get an API key, wire it to your scoring layer, build the dashboards you actually want.
Apache 2.0 · Self-host or cloud · No credit card required