Live-Transcription (STT) with Whisper PoC

nodejs script that enables continuous conversation with voice recognition and tts speaker responses

WhisperMesh is an advanced chatbot that integrates voice and text interactions, delivering personalized responses through LLM models and a sophisticated vector database. Leveraging the RAG framework from Haystack, it ensures engaging, data-driven conversations that adapt to your preferred style.

A small script that types what you say using whisper while holding a hotkey

OpenAI STT custom component for HA

wav2vec를 사용한 STT 기능을 사용하여 음성인식 및 PPT 도우미 기능을 추가

Talk to Rawan voice-to-voice using speech recognition or text-to-speech, with elevenlabs technology and chatgpt on the web.

Electronic Parts Search Assistant - search for any part by text, image/photo or speech. https://parezj.pythonanywhere.com

Witball is a Wit.AI powered Flutter Application that gets data about latest fixtures, current score and also gets the players of your favourite team or any other team you name. Using Witball users can communicate with the Wit.AI bot using a chat interface.

Software converts text into audio and vice versa

PDF Text-to-Speech and Translation

Project exploring IBM-watson speech-to-text and text-to-speech services in python.

gA easy to install/use speach recognition using a webUI with Gradio and faster wisper models (guillaumekln/faster-whisper-large-v2 Is the default)

Your personalized museum tour guide!

AI Chatbot answers students' queries about their college program using Natural Language Processing. It's built with Python's TensorFlow, FastAPI, Uvicorn, React, and Tailwind CSS. The frontend has a modern interface while the backend provides fast and efficient request handling. The chatbot can be customized for specific college requirements

automate telegram account voice to text

This project is a Speech-to-Text and PDF Converter with a modern, visually appealing interface. It allows users to record speech in Hindi or English, converts it to text, and provides options to download the content as a PDF or image. Featuring advanced design, dynamic backgrounds, and accessibility, it’s both functional and engaging.

The automation recognizes face of a human being by facial recognition technique and transfers it to our micro controller which detects the voice command and proceeds with the switching accordingly.

Microservices with HTTP, Triton Inference Server, FastApi and Docker-compose