Token Status

Live health check for every AI vendor CommuniTalk depends on. If one runs out of credits, has a bad API key, or gets rate-limited, it shows red/amber here instead of silently failing on real calls.

Stripe & Billing Status

One click, real Stripe API calls — confirms whether checkout will actually work before a customer hits it. Checks the same secret key and price IDs the real /billing/checkout endpoint uses.

Listening — speech-to-text

How the call page hears each speaker. Cheapest to most premium, left to right. Only Browser (Free) is wired up today — the rest need a real-time audio pipeline that isn't built yet, so they're shown for planning/pricing but can't be turned on yet.

Translation engine

How each utterance gets translated between languages.

Speaking — text-to-speech

How the translated words get spoken back. Cheapest to most premium, left to right. ElevenLabs options automatically fall back to Azure for languages ElevenLabs doesn't support yet (Thai/Cantonese, and Vietnamese on the Multilingual model).

Members

Every account, most recent first — real data straight from the database, not a sample.

Grant Minutes

Comp minutes to any account by email — no Stripe charge. Useful for your own testing, or for comping a real customer goodwill minutes through support. Granted minutes behave exactly like a purchased credit: they never expire and are only spent after the plan's included minutes run out.

Plan Distribution

How many accounts sit on each plan right now, and what each tier contributes to MRR.

Cost → Profit report

Real usage, not estimates — pulled from actual calls and actual vendor tokens/characters billed.

Quality & errors

Latency and failure tracking for every translate-and-speak call, so slowdowns or vendor outages show up fast.

Breakdown by option combo

Which Listening/Speaking combos are actually being used, and how each performs cost-wise.

Sales & ROI

A live planning + reconciliation tool, same shape as admin.voiceleadsai.com's Sales & ROI tab. Real finding worth knowing: the Reports tab's Cost → Profit numbers track AI vendor cost only (Listening + Translation + Speaking) — they do not yet include LiveKit video/bandwidth cost or Stripe's processing fee. This tab adds both on top of your real usage data below (Portfolio Reconciliation), plus a forward-looking planner for pricing decisions (Estimated Profit by Plan) and a customer-goal calculator. Entirely client-side — nothing here changes what's actually live for customers.

Model Settings

Pick which Listening / Translation / Speaking combo to model — defaults to whatever's actually active today. Set your real fully-loaded costs below; every table on this tab uses them.

Break-even minutes — is "unlimited" actually profitable?

Every minute of a call costs real money, so no plan is truly unlimited in cost. This shows how many minutes/month each plan can be used before it stops making a profit, using the fully-loaded per-minute cost from Model Settings above.

Where each cost comes from

So the numbers make sense. Deepgram does speech-to-text only — it turns spoken audio into text. It does not translate, and it does not lower your Cloudflare bill.

Bottom line: Turning Deepgram on adds ~$0.0154/call-min — it is a cost increase, not a saving. Total cost per call goes up slightly, but it's what makes two-way translation actually work on iPhones. Without Deepgram, both sides fall back to the phone's built-in browser speech engine, which is broken on iPhone Safari — so in practice captions only appeared reliably on ONE side (one-way). Deepgram (or another server-side speech-to-text provider) is required for true two-way translation on iPhones. Nothing here lowers the Cloudflare cost, because Cloudflare is already essentially free. Deepgram billing is pay-as-you-go by the minute (no flat monthly fee — the $200 signup credit covers early testing); volume discounts only kick in at much higher usage.

Vendor / Combo Cost Reference

Real rates from the worker's own STT_OPTIONS / TRANSLATION_OPTIONS / TTS_OPTIONS constants (fetched live via /admin/options) — not hand-typed, so this never drifts out of sync with what's actually deployed.

Estimated Profit by Plan

Sell price reflects the discount above. Drag each plan's Low/High minute sliders to bracket real usage — cost, profit, and margin recalculate as ranges, live, as you drag. Fully loaded: Listening + Translation + Speaking + LiveKit + Stripe fee, all deducted before it's called "profit."

Portfolio Reconciliation — real data

Answers "is it really just ~$1/user profit?" using your actual usage this period, not a guess. Left card = what the Reports tab currently shows. Right card = the same real numbers with LiveKit + Stripe fee added on top.

Note: Stripe's fixed $/charge portion isn't modeled here (real transaction count isn't in the usage report) — only the percentage fee is applied to real revenue. The Estimated Profit by Plan table above models the fixed fee precisely, per plan.

Customer Goal Estimator

How many customers, at each plan's midpoint fully-loaded profit, to reach a monthly profit goal.