File size: 1,438 Bytes
a18807a
 
 
 
 
 
 
 
 
 
 
 
c3166eb
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
---
title: Speech2Braille
emoji: 📊
colorFrom: yellow
colorTo: red
sdk: gradio
sdk_version: 3.11.0
app_file: app.py
pinned: false
license: mit
---

# Speech to Braille translator

## Overview
This project can be found [here](https://github.com/Azuremis/speech_to_braille_translator) on my github.

The app currently translates English speech into Braille.

The speech-text translation is done using OpenAI's [Whisper](https://openai.com/blog/whisper/) [model](https://github.com/openai/whisper). The text-braille translation is done using 
the [PyBraille](https://pypi.org/project/pybraille/) library.

## How to use
1. Go to the [demo](https://huggingface.co./spaces/Azuremis/Speech2Braille) page
2. Click the record button and accept the browswer request for microphone access
3. Speak a few sentences then click the stop button when done.
4. Click the submit button to see both the English trascript and the Braille translation


## Disclaimer
This is a portfolio project so I don't advise using it for any serious purpose!

## Credits

Thank you to [Dave](https://twitter.com/azezezaaa), I used his [RememberThis](https://github.com/azezezaaa/rememberthis_OpenAIWhisperHackathon) code to help me grok
how to use the OpenAI's whisper. I also found LabLab's Whisper [guides](https://lablab.ai/tech/whisper) really helpful. 

You can check out the configuration reference [here](https://huggingface.co./docs/hub/spaces-config-reference)