SafeSpeak -A Private English Practice Partner for My Shy Friend
What I Built My friend wants to learn English. He's been wanting to for months. But every time he tries to practice with someone, he freezes. The fear of being judged stops him cold. So, I built him something that doesn't judge. SafeSpeak is a fully local, anonymous English speaking practice partner. It runs entirely on his laptop. No cloud, no accounts, no one listening. What it does: Chat with…
My friend has been eager to learn English for months, yet he would freeze at the thought of conversing with others due to fear of judgment. To help him, I created a solution that offers a safe, judgment-free practice environment, resulting in the development of SafeSpeak. SafeSpeak is an offline, anonymous English practice partner designed to run entirely on a user's laptop, without requiring cloud connections or account creation.
The app facilitates communication with a patient tutor, allowing users to type or speak and receive gentle corrections instead of harsh critiques. It supports multiple languages, including Hindi and Hinglish, with features like voice input for speech-to-text transcription and voice output for natural-sounding responses. The application also enables users to practice various scenarios, such as ordering food, making phone calls, greetings, and giving directions.
SafeSpeak's primary purpose is to provide an environment where users can make mistakes without any consequences, encouraging a learning experience without the pressure of judgment. The code is open-source and available on GitHub, and it can be set up locally on any device without the need for an internet connection. This ensures privacy and freedom for users to practice at any time, even during late hours or in areas with poor connectivity.
SafeSpeak is built utilizing open-source AI technologies, including Gemma 3 (1B) via Ollama, Whisper for speech-to-text, and Piper for text-to-speech. The entire application is designed to operate offline, avoiding cloud APIs, API keys, and any form of telemetry. This approach not only addresses the user's privacy concerns but also makes the app accessible to anyone, regardless of their internet availability.
Prize entries for this project include the category "Best Use of Gemma," highlighting the app's reliance on the open-weight model. The creation of SafeSpeak was motivated by a desire to offer a friend a supportive platform for practicing spoken English, catering to the needs of shy beginners in a respectful and open innovation manner.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
