AbdelStark

AbdelStark / mistral-parler

Public

🇫🇷 parler: Multilingual voice intelligence built on Mistral Voxtral model — decision logs from French/English meetings

french-tech mistral-ai open-models speech-to-text
12
1
100% credibility
Found Apr 14, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

parler is an open-source Python application that processes meeting audio files to generate structured decision logs including decisions, commitments, open questions, and rejections using AI transcription and extraction.

How It Works

1
📰 Discover parler

You hear about a helpful tool that turns your messy meeting recordings into clear lists of decisions and next steps.

2
📦 Set it up easily

Download and prepare it on your computer in moments, ready for your audio files.

3
🎙️ Choose your recording

Pick an audio file from a meeting, add names of people talking and the date.

4
🚀 Hit go and watch

Press run to see it listen, understand speakers, and pull out key decisions live.

5
📋 Review the magic

Open beautiful reports with decisions, promises, questions, and who's doing what.

✅ Share and act

Your team gets crisp notes to move forward, saving hours of manual work.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 12 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is mistral-parler?

Mistral-parler processes meeting audio into structured decision logs, transcribing French/English conversations with Mistral's Voxtral model and extracting decisions, commitments, questions, and rejections via Mistral LLMs. Drop in MP3/WAV files via CLI (`parler process meeting.mp3 --lang fr,en --participant Pierre`), get Markdown/HTML/JSON outputs with timestamps and owners—perfect for turning chaotic talks into actionable intelligence. Python-based with TUI dashboard, local inference, and caching for repeat runs.

Why is it gaining traction?

It nails multilingual French/English meetings out-of-box, with cost estimates, speaker attribution, and resumable checkpoints to avoid API waste—unlike generic STT tools lacking decision extraction. The Mistral integration feels native (parler avec Mistral AI vibes), plus verbose logs and roster for known participants make iteration fast. Devs dig the uv sync setup and TUI for real-time pipeline monitoring.

Who should use this?

PMs or engineering managers summarizing bilingual standups/client calls into Jira/Slack exports, tired of manual log-parsing. French/English teams building decision intelligence around Mistral models, or anyone testing parler tts multilingual for meeting communication github flows.

Verdict

Promising alpha for Mistral fans (11 stars, 1.0% credibility score), with polished docs, CLI/TUI, and strong tests—but too green for heavy production. Prototype your multilingual logs here; watch for stability as it grows.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.