Real-time AI coaching for sales and negotiation calls.
Note-takers tell you what happened in a call after it's over — when it's too late to change the outcome. Parley coaches you while you can still act: it listens to the meeting, flags what just happened, suggests what to say next, and tracks where the deal stands, live.
- 🎙️ Coached, live — a real-time feed of alerts and suggested replies, plus an intelligence board that extracts the state of the deal as you talk.
- 🌐 In any language — speak your language; the meeting hears the other one, through Parley's own virtual microphone.
- 📼 Debriefed, after — commitments, missed moments, and a delivery scorecard for the next round.
- 🗂️ Remembered, between — an account workspace that turns every call into durable intelligence: people, deal threads, and what you've learned about each side.
Local-first, bring your own keys. Audio and transcripts go directly to the STT and LLM providers you configure (Claude, OpenAI, Gemini, Soniox, Deepgram, …). No Pathors proxy, no telemetry, everything stored on your machine.
Note
macOS only (for now). Parley uses a Core Audio process tap for system-audio capture and ships a CoreAudio virtual-microphone driver.
Download the latest build from the Releases page, open the .dmg, and drag Parley into Applications. Builds are signed and notarized — no Gatekeeper hoops.
Then paste your API keys in Settings: one STT provider for transcription, one LLM for coaching. For live translation, add a Gemini key and install the Parley Microphone with one click.
Parley captures both sides — your mic and the meeting's system audio — and transcribes them live, diarized as me / them.
On top of the transcript, the coach feed raises evaluation alerts (negotiation risk, qualification gaps, red flags, or your own rubric), each with a drill-down into how to reply. Ask it anything from the input bar and get answers grounded in the conversation so far.
While the feed tells you what just happened, the intelligence board tells you where the deal stands. Tell Parley what kind of meeting this is and it extracts the structured state as you talk:
| Meeting type | What the board tracks |
|---|---|
| ⚖️ Negotiation | Every number said and by whom, the concession ratio between sides, agreed vs. open terms |
| 🤝 Sales | Budget / timeline / decision-maker signals, objections answered vs. still open, commitments, competitor mentions |
| 🚀 Partnership | A live they-have × they-need map, concrete mutual-leverage proposals, give/get balance |
Plus an auto-checked agenda and live delivery nudges (pace, pitch, pauses) measured on your mic only.
Speak Mandarin; the meeting hears English — live, in 70+ languages, with your intonation preserved.
Flip one switch and your side of the call runs through Gemini speech-to-speech translation, out through the Parley Microphone — a signed CoreAudio virtual mic that installs with one click. Select it as your mic in Google Meet, Zoom, or Teams, and the translation is what they hear. The other side stays on your regular transcription pipeline, and the bilingual transcript feeds the coach like any other meeting.
A slim interpreter strip shows the live original → translation line, a running cost ticker, and a pause switch; a standalone quick interpreter window covers in-person conversations.
Stopping a meeting lands on its debrief. Any recording — just finished, from history, or dragged in as an audio file — opens in two views:
- Report — one scroll: the debrief with clickable timestamps, both sides' commitments, action items, the intelligence board's final state, and your delivery scorecard (measured pace, talk share, filler sounds).
- Replay — the full player: scrub to any moment and re-run the analysis as of that point, review the other side's moves and your missed moments side-by-side with the transcript, and ask anything about the call from a drawer that follows you across tabs.
Everything generates once and saves with the recording — reopen it a month later and the whole report loads instantly, no extra LLM calls. Plus LLM speaker re-attribution fixes diarization drift by conversational context.
Deals span many calls; Parley remembers what each one taught you. The account workspace keeps a living dossier per company:
- Intel cards — atomic facts across nine categories (stances, leverage, goals, risks, red lines, competitors, next moves, open questions, relationships), each with its source quote, confidence level, and freshness. Conflicting intel surfaces in a triage queue instead of silently overwriting.
- Deal threads — each negotiation or opportunity tracked on a pipeline (discovery → demo → proposal → negotiation → closing) with stage guides for what to collect next.
- Stakeholder map — who's the economic buyer, who's your champion, who blocks, and who influences whom.
- The loop closes both ways — after a call, one review pass files what the meeting revealed into the dossier; before the next one, a single click assembles a briefing and loads the stage's open questions into your agenda. Red-line cards automatically arm as live alerts when a meeting with that company starts.
Conversation content is sensitive, so Parley runs straight from your machine:
- Direct connections — audio and transcripts go only to the providers you configure, under your own keys.
- Local storage — recordings, transcripts, and templates stay in your local app directory.
- No telemetry — nothing tracked, collected, or uploaded.
- Voice typing — system-wide push-to-talk dictation in any app, using your configured STT provider. Hold a key (default Option+Space), speak, release — the text pastes into the frontmost app.
- Built-in MCP server — connect Claude (or any MCP client) to the live meeting: read the transcript, manage agenda TODOs, edit the timeline analysis.
- Traditional Chinese — full zh-TW UI and on-the-fly conversion of transcribed text.
Requires Rust (stable) and Bun (or Node.js):
git clone https://github.com/pathorsAI/parley.git
cd parley
bun install
bun run tauri dev(optional) Build and install the virtual-microphone driver for translation into meetings:
cd virtual-mic && ./build.sh && ./install-dev.shContributions are welcome! See CONTRIBUTING.md for how to report bugs, suggest features, and submit pull requests.
Licensed under the Apache License 2.0. Copyright 2026 Pathors AI.