The self-learning voice for finance
Voice agents trained on your calls, checked on every turn, improving every week under your rules — in the languages your customers speak.
Trusted by: IndusInd Nippon Life Insurance, HDFC Bank, TeamLease RegTech, Teachmint, iTuring, Sheopals, Conscious Chemist, HealthFab, Mopid.
It speaks like your best agent.
Fine-tuned on your own recordings — accent, pace, numbers and dates the way your customers say them.
It checks every turn.
Grounded in account data, your script and your regulator's rules before a word is spoken.
It gets better every week.
Graded calls become the next release — regression-tested, human-approved, one-click rollback.
It stays inside your walls.
Your model, your keys, your region. India cloud, private VPC or on-prem GPUs.
Most voice bots are as good on day 300 as they were on day 1. Ours run a governed weekly learning cycle on your own calls.
Every turn checked before it is spoken.
- 00:04Namaste, main Riya bol rahi hoon, [Lender] ki taraf se. Kya main Sunil ji se baat kar rahi hoon?नमस्ते, मैं रिया बोल रही हूँ… क्या मैं सुनील जी से बात कर रही हूँ?
- 00:11Haan bol raha hoon. Kya hua?
- 00:14Aapki 5 tareekh ki EMI ka payment abhi tak nahi aaya. Kya aap aaj UPI se pay kar payenge, ya main payment link WhatsApp par bhej doon?आपकी 5 तारीख की EMI का पेमेंट अभी तक नहीं आया…
- 00:27Salary Friday ko aayegi, tab kar dunga.
- 00:31Theek hai Sunil ji, main Friday, 18 September ka promise note kar rahi hoon. Link bhi bhej rahi hoon. Shukriya.ठीक है, मैं शुक्रवार 18 सितंबर का प्रॉमिस नोट कर रही हूँ।
Every call makes the next one better.
A turn check inside a call grade inside a weekly release inside a portfolio — four loops, each feeding the one outside it.
The city calls
Recordings, transcripts, outcomes, human takeovers and compliance flags stream in as signals — 100% of calls, not a sample.
48,210 signals · 7 days illustrativeOne hundred calls, graded
Grounding, flow, compliance, tone. Failures are clustered: "repeated the EMI amount", "missed already-paid", "asked branch-or-link twice".
100% graded · 14 clusters illustrativeRedaction, then selection
Aadhaar, PAN, card and phone numbers masked in text and audio. Your best agents' calls and the corrected failures become the training set — approved by your reviewer.
PII 100% · 3,120 examples illustrativeYour weights
Speech and language models fine-tuned per tenant. No shared brain. The weights are yours, stored in your environment for VPC and on-prem.
candidate v42 · your tenant illustrativeTwelve hundred and forty scenarios
The candidate replays the golden set — barge-in, angry customer, already-paid, wrong number — with Hindi/English parity and frozen-line checks. Any drop blocks release.
1,240 scenarios · 0 red illustrativeA human signs
Named approval from your QA or compliance owner, staged rollout 5% → 25% → 100%, automatic rollback on a guardrail breach.
5% → 25% → 100% · 1-click back illustrativeSix weeks, the dip left in
Containment, promise-to-pay, right-party contact, complaints and cost per resolved call — tracked per release, reported every morning. Then the next week begins.
PTP +8.5 pts over 6 weeks illustrativeNext week's calls feed the next cycle.
Client memory carries over: scripts, glossary, frozen lines, escalation matrix.
The models that grade the models.
A second model, trained on your QA team's labels, scores every call on grounding, flow, compliance and persona. It replays 1,240 scenarios against every candidate — and any regression blocks the release. Your reviewers recalibrate it weekly.
Governed. Personal. Remembering. On the floor.
Every idea in voice AI, translated into what it means for a regulated call — and wired into the same engine, learning cycle and evaluation models.
What it may see, say and do.
A policy engine between the agent and every field, line and action — with a "why" on every turn.
Governed voice agentsEvery call teaches the next.
Turn, call, week, portfolio — four loops, each feeding the one outside it, under your rules.
Recursive learning loopsModels that grade the models.
A second model scores every call; a simulator replays 1,240 scenarios before anything ships.
Evaluation modelsThe customer never repeats themselves.
Four tiers of memory, redacted before storage, forgotten on request, with provenance on every fact.
Context & memoryPersonal, never improvised.
Language, pace, tone, timing and offers — drawn only from governed fields, never inferred.
PersonalizationOn the floor, not instead of it.
Warm transfers with a full brief, whisper to human agents, after-call work done for them.
AI coworkersYour best agent. Your customers. Your journey.
Three twins, each trained, tested and governed like any release.
Digital twinsHired like people.
A role, a manager, KPIs on outcomes, a weekly review, a permission set — and an ID badge on every call.
AI employeesInsurance. Banking. Lending.
Collections, renewals, KYC, lead qualification and service — in the languages your customers actually speak.
Three ways in. One governed engine.
Start with one journey
One use case, one language, your scripts. Live calls in week 3, learning cycle from week 4, a decision by week 6.
How we workBuild on the engine
Model Studio fine-tunes speech and reasoning on your recordings. Telephony, WhatsApp, SMS and chat from one brain.
Explore the platformOwn the whole stack
Private VPC or on-prem GPUs, air-gapped option, your keys. Security reviewed against RBI, IRDAI and DPDP.
Deployment optionsFrom an IIT lab to live calls in five countries.
FormantAI started as speech and reasoning research — the hard problem of understanding Hindi, Hinglish and Tamil the way people actually speak on a phone line. It became a company when the first bank asked us to run it inside their walls.
It began in the lab
Speech and language research at IIT, on code-mixed Indian speech over real telephony — the audio most models were never trained on.
Then a first live journey
A regulated lender needed a voice agent that could be audited, not just demoed. Every turn checked, every line frozen where the regulator demands it.
Then it learned
Graded calls became weekly releases — loops inside loops, judged by evaluation models we built before the learner. Better in month six than in week one.
Now live in five countries
India, UAE, USA, Vietnam and South Africa. Part of Google for Startups. Trusted by IndusInd Nippon Life Insurance, HDFC Bank and TeamLease RegTech.