stt-api

Examples and guides for using the Smallest AI

It's built by a developer, for developers who want a tool that gets out of the way.

RaDUO turns live radio broadcasts into multilingual STT datasets for ASR research. It records stations in parallel, splits streams into clips, transcribes audio, extracts quality features, and stores searchable metadata. Includes 16 kHz mono WAVs, transcripts, station info, languages, and optional word timings.

Relay Time is a modern Android app built with Jetpack Compose that transforms speech into text and text into speech using Google’s STT and TTS APIs. Designed to run smoothly on any screen size, it features a clean, responsive UI with simple navigation between voice input and text playback—making communication effortless in both directions.

Golang client library for Gradium.AI TTS/STT API

Realtime STT/TTS library allowing provider agnostic implementation

AI-powered personal health companion app — Flutter + FastAPI. Features a RAG-based medical assistant, consultation tracker with OCR document uploads, voice emotional buddy (Orbz), mood analytics, and PDF report export. Backed by Supabase, Groq, and Qdrant.

Speech-To-Text (STT) project

an extremely fast qwen3-asr-0.6B model on CPU for hermes agents and others.

Browser-based AI Teleprompter for public speakers

easy api stt for whisper.cpp .Same ollama but use whisper.cpp

Meet VoiceFlow 🎙️🔊, your production-ready microservices platform for all things AI speech! It's designed to make high-performance voice processing a breeze, letting you effortlessly transcribe audio to text and convert text into natural-sounding speech. 🚀

Prototyping calling local STT engine from a Tauri2 app

a local, privacy‑first voice interface that lets you speak to an AI companion. It captures your speech, converts it to text via STT, sends it to an LLM (e.g. via your local backend), then converts the response back to speech (TTS) and plays it — all from front‑ to back‑end.

Web-based control panel for self-hosted LLM inference on NVIDIA GPUs — vLLM, MIG, LiteLLM gateway, HTTPS, monitoring.

Add to my main repository

FocusMate, an ESP32-based AI productivity assistant designed to support focused work sessions through conversational interaction instead of traditional screen-based tools. The system integrates Gemini-powered intelligence with cloud-based speech-to-text and text-to-speech APIs to enable real-time voice interaction.

data science contest

Fish Audio CPP client for TTS and ASR

speech to text on CUDA