Skip to content

Web app that transcribes speech to text. In this case, You Tube audio and uploaded audio files to text.

License

Notifications You must be signed in to change notification settings

rmerzouki/speech-to-text

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

47 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Transcription Application

1. Create an AssemblyAI free account

Create your free AssemblyAI Account & obtain your API Key.

2. Running the transcription application as a Streamlit app

To recreate this web app on your own computer, do the following.

Create conda environment (Optional)

Firstly, we will create a conda environment called transcriber_env

conda create -n transcription_env python=3.7.9

Secondly, we will login to the transcription_env environment

conda activate transcription_env

Download GitHub repo

git clone https://github.com/rmerzouki/speech-to-text.git

Pip install libraries

pip install -r requirements.txt

Launch the app

streamlit run audio_to_text_2.py

3. Deployment on Streamlit Cloud

How to deploy on Streamlit Cloud your Speech-to-Text App in Python using AssemblyAI

How to deploy on Streamlit Cloud your Speech-to-Text App in Python using AssemblyAI

 Steps to deploy the app on Streamlit :

    1 - Run your app locally, click on the Streamlit menu at the top right of the page and select "Deploy this app".
    2 - Apps are deployed directly from their Github repo. Enter the location of your app and the settings i.e. secrets, in our specific case the api key i.e. api_key = 'replace_with_your_AssemblyAI API_key' 
    3 - Save & Hit Deploy!
    4 - Once the application is deployed, share your web application path with the audience.

Streamlit Secrets Management allows you to store secrets securely and access them in your Streamlit app as environment variables. For further information on this topic, check out the following documentation :

 https://docs.streamlit.io/streamlit-cloud/get-started/deploy-an-app/connect-to-data-sources/secrets-management


 For your information, my web app is deployed on Streamlit and it is accessible through the following path :
 
https://share.streamlit.io/rmerzouki/speech-to-text/main/audio_to_text_2.py

References

* [1]  Chanin Nantasenamat https://youtu.be/NNq_XBVk30w

Additional notes

A bug in pytube led me to temporarily use git+https://github.com/rmerzouki/pytube ; this bug is now fixed with pytube version 12.0.0. 
A new bug in pytube led me to temporarily use git+https://github.com/jmsmg/pytube ; this bug is now fixed with pytube version 12.1.0 released in May 9, 2022 and available in https://pypi.org/project/pytube/
 
A new bug in pytube led me to use git+https://github.com/jmsmg/pytube as of April 26th 2022.

About

Web app that transcribes speech to text. In this case, You Tube audio and uploaded audio files to text.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages