README.md

Speech Recognition Transcriber: speechRecognitionTranscriber (Python API)

• Introduction
• Use
• Requirements
• Status
• Related projects

Introduction

speechRecognitionTranscriber module use Google Speech API in python. This module performs speech recognition and converts to text. Admits video and audio files to be transcribed. Use ffmpeg to convert video and audio files to .wav to be recognized in Google Speech API. Also use fragments division based on silence.

Documentation available on docs.

Running software

speechRecognitionTranscriber requires video/audio like input. The process to running the program:

1. Execute programs/speechRecognitionTranscriber.py, to start de program.

python speechRecognitionTranscriber.py

2. Introduce your file path.

yourfile.extension

NOTE:

• Transcribed text is saved in transcribedText.txt.
• Transcribed text is saved in transcribedText.pdf.
• Audio fragments are saved in /fragments.
• Converted source is saved as convertedFile.wav.

Temporal files like cconvertedFile.wav and /fragments are deleted when program ends.

Requirements

speechRecognitionTranscriber requires:

• Install pip
• Install SpeechRecognition:

pip install SpeechRecognition

• Install fpdf:

pip install fpdf

• Install pydub

pip install pydub

• Install ffmpeg

Linux

sudo apt-get install ffmpeg

Microsoft Windows Download binaries and set path in system variables.

Tested on: windows 10,ubuntu 14.04, ubuntu 16.04, ubuntu 18.04, lubuntu 18.04 and raspbian.

Status

Related projects

• realpython: python speech recognition
• python basics: transcribe audio