Volavi-FSTT (Finnish Speech-To-Text)

This is a personal project I created for myself.

Volavi-FSTT is a simple desktop application I built to help me transcribe spoken thoughts — mainly in Finnish or English — into written text. I use this tool for therapy-like self-reflection: I talk out loud, record or save the audio, and then use this program to transcribe it. I feed that transcript into another AI system that helps me plan, reflect, and process things.

I made this to run entirely on my own hardware, for free, with no cloud, no accounts, and no data leaving my machine.

I'm sharing it on GitHub as freeware, for anyone who might benefit from something similar. But just to be clear: this is not a commercial product, I'm not looking for fame or money, and I don’t offer any support or guarantees.

---

🧠 Why I Made This

I needed something fast, accurate, and private to convert speech into text for my own use. Nothing online fit my needs exactly — I wanted:

• Whisper’s high accuracy
• A user interface that just works
• Full control over files and processing
• No subscriptions, no APIs, no cloud

So I built it.

---

🎯 Features

• 🎙️ Transcribes audio files (MP3, WAV, FLAC, etc.) using OpenAI Whisper
• 🌍 Supports multiple languages — I mostly use Finnish and English, but others are available
• 📂 File browser to load media from any folder
• 📝 Auto-saves transcripts as .txt files
• 📋 View, copy, and open transcripts from inside the app
• 🧘 Built for self-reflection, journaling, or therapeutic processing

---

🔧 Requirements

You’ll need:

• Python 3.8 or newer
• Whisper installed: pip install git+https://github.com/openai/whisper.git
• FFmpeg installed (make sure it's in your PATH)
• Tkinter (usually included with Python)

---

🚀 How to Use

1. Put your audio files in the media/ folder (or choose another folder in the app)

2. Launch the app:

   python volavi-fstt.py

3. Select your preferred Whisper model and language

4. Choose audio files and click “Transcribe Selected”

5. View the transcript in the right panel

Transcripts are saved automatically to:

media/transcripts/{file_stem}_{hash}/{file_stem}.txt

---

📂 Project Structure

volavi-fstt/
├── volavi-fstt.py       # The main app file
├── media/               # Default folder for audio files
│   └── transcripts/     # Transcriptions are saved here

---

🛑 Disclaimer

This is freeware. I built it for my personal use and make it available as-is. I’m not responsible for:

• Any issues, bugs, or incorrect output
• Loss of data
• How you choose to use it

I don’t offer support, updates, or guarantees. Use at your own discretion.

---

🙌 Final Note

This tool is part of how I clear my head and process thoughts. If it helps you too, great. You're free to use, modify, and adapt it for personal use.

— Developed with care, by someone who just needed a quiet, local solution.

---