Fluid Enhanced Dynamic Text-to-Speech
A powerful Streamlit app that combines an audio transcriber, a read-aloud voice reader, and a grammar checker — all powered by deterministic algorithms. No neural networks. No cloud APIs. No LLMs. Just fast, private, reliable tools that work.
FED TTS brings together three essential text tools in a single, beautiful Streamlit interface. Each one works completely offline using proven, deterministic algorithms.
Upload your audio files (MP3, WAV, M4A, FLAC, OGG) and transcribe them manually with a built-in player featuring rewind, fast-forward, and speed controls. Your transcript persists across page interactions.
Have any text read aloud using your browser's native SpeechSynthesis API. Uses your operating system's built-in voices — no cloud TTS services, no API keys, no subscriptions. Just pure, local text-to-speech.
A Grammarly-style checker powered by dictionary-based spell checking (pyspellchecker) and regex grammar rules. Catches common mistakes like "would of", passive voice, double spaces, and misspelled words — all deterministically.
FED TTS is built on a radical idea: powerful tools don't need artificial intelligence. Every feature works with deterministic algorithms that are fast, private, and reliable.
Experience FED TTS right here in your browser. Test the grammar checker, hear the text-to-speech, and explore the transcription workflow — all without installing anything.
Click "Check Grammar & Spelling" to see results.
SpeechSynthesis API.
The voices you hear are provided by your operating system — no cloud, no AI.
Try adjusting the rate, pitch, and volume sliders!
Click to upload or drag & drop
MP3 · WAV · M4A · FLAC · OGG
No microservices, no message queues, no neural networks. Just a clean Streamlit app with three deterministic modules and your browser's native APIs.
Choose your preferred installation method. FED TTS runs on any system with Python 3.9+.
Clone and run locally:
# Clone the repo
git clone https://github.com/fed-tts/fed-tts.git
cd fed-tts
# Install & run
pip install -e .
streamlit run app.py
Install as a package:
# Install from PyPI
pip install fed-tts
# Launch the app
streamlit run app.py
# Or use as a library
python -c "import fed_tts"
Run in a container:
# Build the image
docker build -t fed-tts .
# Run on port 8501
docker run -p 8501:8501 fed-tts
# Open localhost:8501
Deploy for free:
# 1. Fork the repo on GitHub
# 2. Go to
# share.streamlit.io
# 3. Connect & deploy
# Main file: app.py
FED TTS is actively developed. Here's where we've been and where we're going.
File upload, manual transcription, browser TTS, dictionary spell check, 12+ grammar rules, Ko-fi support button, full CI/CD pipeline, 19 unit tests, MIT license.
Dark mode toggle, custom HTML5 audio player with keyboard shortcuts, TXT/PDF export, reading time estimate, search & replace within transcript, auto-save drafts.
Plugin system for custom grammar rules, audio waveform visualization, large file optimization, streaming audio playback, test coverage >90%.
PyInstaller desktop executable, multiple language support, batch processing, browser extension, mobile companion app, auto-update mechanism.
Join the FED TTS community. Star the repo, try the demo, or support the project on Ko-fi.