Skip to main content

A project to use Sentence Transformers and embeddings to make a pocket search engine

Project description

Project: Oráculo

Oráculo is a versatile CLI and WebApp application developed for transcription of audios and semantic search. It leverages Sentence Transformers and embeddings to create a compact search engine that aids in retrieving and organizing important information from a collection of documents.

This application is particularly useful for professionals dealing with substantial amounts of audio data and requiring an efficient system to transcribe and conduct semantic search operations on the data.

Features:

  • Audio Transcription: Oráculo can transcribe audio files. You can transcribe a single file or bulk transcribe a folder.
  • Semantic Search: A web app to perform semantic searches on the transcribed audio data.

Requirements:

:warning: IMPORTANT :warning: In order to run Oráculo, you need to have the following requirements installed on your machine:

  • Python 3.10
  • FFmpeg
  • Git

Installation:

You can install Oráculo with pip:

pip install oraculo

Setup:

:warning::warning:Warning:warning::warning:: The following steps are required to run Oráculo. Please follow the steps carefully.

Initialize the Oráculo application with the following command:

oraculo init

You will be prompted to enter the following information:

Information Description
ChromaDB Persist Directory The directory where the ChromaDB will be stored. This is important to store vector embeddings of text
ChromaDB Implementation Defaults to duckdb+parquet. For more implementations, please refer to Source Code

Whenever you want to change the config file, just run the same command again.

Usage:

Semantic Search:

To start the Semantic Search Application, use the following command:

oraculo webapp

Single File Transcription:

To initiate a transcription for a single file:

oraculo transcribe

Multiple File Transcription:

To initiate bulk transcription for a folder:

oraculo bulk-transcribe

to transcribe youtube videos:

YouTube Video Transcription:

oraculo transcribe-yt

Help:

If you need help with the commands, use the following command:

oraculo --help

About

The development of Oráculo is aimed at information retrieval capabilities for businesses and individual users. Please feel free to reach out with any feedback or suggestions to improve Oráculo further.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

oraculo-0.1.14.tar.gz (8.2 kB view hashes)

Uploaded Source

Built Distribution

oraculo-0.1.14-py3-none-any.whl (9.4 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page