Taveni · French software built by Hyle Labs

Your meetings, structured as decisions and actions to review.

After a meeting, the essentials get lost: who decided what, by when, in which tool. From the recording, Taveni proposes decisions and actions, with an owner and due date; your team checks every proposal before anything is sent.

Evaluation on a real process · Scoped engagement · No commitment

  1. The meeting — The sentence where the decision is made.
  2. The proposal — Taveni drafts a decision from it, with its confidence level.
  3. Your review — You correct, validate, or reject it before anything is sent.
Illustrative example · fictional data See an actual screenshot
  • Source excerpt attached to each proposal, when one exists
  • Owner and due date proposed when the context allows
  • One meeting is enough to start
The problem

Your best decisions are made out loud. Then they vanish.

Automatic summary tools produce recaps that nobody reopens. The value is elsewhere: in the decisions made out loud, which must be found, reviewed, and executed.

  1. The decision evaporates.

    Budget, priority, an exception granted to a customer: the decision is made out loud, then lost inside a long transcript.

  2. The next step is too vague.

    "I'll handle it" is not clear enough to create a reliable task in a project or support tool.

  3. Nothing reaches the tools.

    The recap exists, but the action reaches neither the CRM, support desk, nor project tool — and nobody controls who can see what.

Two angles often forgotten: access rights and the AI bill
  • Sensitive information — sharing everything with an assistant is not acceptable for every account: access rights must follow the decision.
  • The AI bill — the cheaper AI gets, the more is sent without filtering, and you pay for noise.

Taveni is designed to turn meeting material into sourced operational memory: decisions and actions to review before any handoff.

Where to start

Start with the meeting that demands the most manual follow-up.

Pick one recurring meeting and the place where its follow-ups are handled: that is where you judge whether Taveni helps.

TeamTypical meetingWhat Taveni does with it
LeadershipCommittees and trade-offsFind why a priority, budget, or date was approved, with the excerpt and expected next steps.
Sales & customer successCustomer meetingsCommitments made in customer meetings become actions to review, then to hand off to the CRM, support, or a follow-up.
Product & opsOperational decisionsWhat was decided across teams stays visible, including actions without a clear owner.
How it works

Four steps. Approval before sending.

Taveni prepares proposals; your team reviews and approves them before anything is sent.

Capture.

Import the recording of a meeting or call, or start a live session.

Meeting and context together

Qualify.

Give Taveni your team's context: roles, projects, vocabulary, access rules.

Proposals to review

Review.

Confirm important decisions and actions or keep incomplete items pending.

Content approved by your team

Hand off.

After approval, the action reaches your tool: connector or documented export.

Controlled sending

Taveni interface: actions extracted from a sales review, with priority, confidence level, due date, source excerpt, and a destination to connect.

Actual Taveni screenshot

Qualified actions, waiting for your review.

  • Due date, priority, confidenceProposed from the meeting, editable before validation.
  • Quoted excerptEach action keeps the meeting excerpt it comes from, when one exists.
  • Destination to connectNothing is sent until the action is approved and its destination connected.
See how it works in detail
  • Capture — the audio is transcribed and timestamped: every passage is linked to the moment it was said.
  • Qualification — Taveni proposes decisions, actions, and risks, with owner, deadline, access rule, and destination.
  • Review — your teams confirm, complete, or reject; reviewing sends nothing.
  • Handoff — only after explicit approval, through a connector or a documented export.
Taveni interface: timestamped transcript of a sales review — the decision “We approve the Q3 budget at €180,000” and the proposed action with its CRM target.
Taveni — decision linked to its evidence
Taveni interface: the review queue of decisions, commitments and risks extracted from a meeting, with confidence level and status.
Taveni — human review of decisions and actions
Taveni transcript of a sales review, timestamped per speaker, with detected and validated entities.
Taveni — timestamped transcript and qualified entities
Taveni dialog to correct an item before validating it.
Taveni — human correction before validation
Taveni interface, Integrations page: the ClickUp card shows Connected, with a chosen destination; every outgoing task requires exact approval.
Taveni — ClickUp connection with a controlled destination
Taveni dialog: review of the ClickUp task before creation, with the destination list and the reminder that any change requires a new preview.
Taveni — human approval before sending to ClickUp
Taveni interface: a qualified action carries the ClickUp mark of its created task, while the other actions remain to be sent.
Taveni — action linked to its ClickUp task
The difference

Taveni starts where automatic notes stop.

Automatic notes (AI-generated transcripts and summaries) tell you what was said. Taveni aims to structure what deserves operational review: source, owner, status, access rule, and proposed destination.

Why sending everything to AI gets expensive.

As AI gets cheaper, more unfiltered context gets sent back through it. Taveni is designed to separate raw verbatim from qualified information.

The Jevons paradox, in a nutshell

When a resource gets cheaper, its use often increases. For AI, that makes calls and processed context worth measuring rather than assuming savings.

TopicClassic AI notes toolTaveni angle
SummaryProduces a readable summary, often detached from execution.Isolates the decisions, actions, risks, and evidence worth keeping.
QualificationDetects tasks or next steps from the conversation.Requires source, owner, access right, review status, and destination.
ExecutionSends notes or actions based on available integrations.Prepares the handoff, after validation only.
ControlControls vary by product and configuration.Requires access, destination, retention, and cost to be scoped before the handoff.
The cost of AI

Measure the context that is actually useful before sending it to the AI.

Every AI processing run has a cost, which depends on the model and the volume sent (the "context"). The evaluation compares your starting point with the context actually processed.

How context gets qualified

Before extraction, Taveni relies on qualified context (see the example, fictional data): the meeting profile, business vocabulary, and your organization's projects and clients.

context-example.json
{
  "meeting_title": "Q3 sales review",
  "participants": [
    { "name": "Nathalie Dupont",
      "role": "CFO" },
    { "name": "Karim Benali",
      "role": "Sales Lead" }
  ],
  "user_profile": {
    "company": "Example Ltd",
    "job_title": "Finance lead"
  },
  "vocabulary": [
    "ARR", "renewal", "Q3 committee"
  ],
  "org_projects": ["Q3 Expansion"],
  "org_clients": ["Acme Corp"],
  "org_products": ["Taveni"],
  "output_instruction": "Prioritize decisions, actions, and evidence.",
  "routing_rules": "Do not send without owner, source, or right"
}
whodirectory and roles clarified
whatprojects, clients, vocabulary retained
measureprocessed context compared with baseline
Trust & sovereignty

Delegate the task, not the learning.

The AI does the sorting; what your team teaches it by correcting — decisions, access rules, vocabulary — stays yours. A proposal only counts once a human has reviewed it.

Detected phraseQuestion to resolveNext step
"We approve the Q3 budget"Can we find the excerpt that supports this decision?to verify
"I'll handle it Monday"Are the owner, deadline, and subject clear enough?to complete
"Confidential account topic"Should this information remain limited to certain people?to confirm

If the AI model changes, the evaluation verifies which decisions, corrections, and rules remain exportable and reusable.

InformationOriginState
decisionmeeting · 00:17:42 · oral proofvalidated
commitmentspeaker · excerpt · confidenceneeds review
riskaccess rules · confidentialityrestricted
  • Sourced decisions. The link between a decision and its evidence is checked before any sharing.
  • Share the minimum useful context. The scope states which information each tool or assistant may receive.
  • Sensitive data. Access, retention, and sending rules are documented before starting.
  • Learning loop. The evaluation measures whether validations, rejections, and corrections improve later suggestions.
Hyle Labs

AI products that turn speech into capital.

Hyle Labs designs and publishes Taveni. We build products for teams whose decisions move between conversations, tools, and assistants.

“I created Taveni because useful decisions were buried in meeting notes. The goal is to make them easier to find and review, without ever taking the decision away from people.”

Adryan Perez — founder, Hyle Labs · Lyon

  • French company, based in Lyon
  • Taveni in scoped evaluation
  • Privacy policy published
  • No third-party trackers on this site
Straight answers

What teams ask before evaluating.

Do we need to change tools to use Taveni?

No: Taveni is a web application used alongside your current tools; the evaluation defines a first handoff, through a connector or a documented export.

Where does the data go? (GDPR)

It depends on the agreed scope: before the evaluation, we document the data involved, the services used (including the AI model provider), access, and retention.

What happens if we switch AI models?

The evaluation documents what remains exportable and reusable if the model changes, according to the formats and terms agreed during scoping.

How much does it cost?

Taveni is not self-service: you evaluate it on one of your recurring meetings. Scope, success criteria, resources, and commercial terms are defined before anything starts.

Which tools do you connect?

A CRM, support, or project-management tool (ClickUp in the screenshots on this page). Connector availability is checked during scoping; otherwise, a documented export.

Talk with us

Evaluate Taveni on one of your meetings.

Describe the meeting, the target tool, your GDPR constraints, and your success criteria. We reply with a scope, measurement criteria, and the conditions of the evaluation, which then runs on your real meetings.

contact@hyle-labs.net

Evaluation on a real process · Scoped engagement · No commitment

What an evaluation measures

  1. Search timeThe time to find a decision and its source.
  2. Manual re-entryWhat still has to be copied by hand.
  3. Errors found before sharingErrors caught before sharing or sending.
  4. Observed AI costThe AI cost on the evaluated sample, measured rather than estimated.
Request an evaluation