Release LangChain Python based samples to illustrate concepts on Intel hardware. There are currently three samples for release: Audio Chapterization, Audio/Text QnA with/without TTS, and Video Summarization.
Audio Chapterization - Demonstrates a pipeline which automatically chapterizes long text/content from a provided audio context. The primary components utilize OpenVINO™ in LangChain* for audio-speech-recognition, embeddings generation, K-means clustering, and LLM chapterization.
Audio/Text QnA - Demonstrates a pipeline which performs QnA using audio or text with RAG. The primary components utilize OpenVINO™ in LangChain for audio-speech-recognition, LLM text generation/response, and text-to-speech.
Video Summarization - Summarize Videos Using OpenVINO™ with Langchain