https://github.com/kefniark/voice-your-choice

Small Openhack prototype about Web Speech API
https://github.com/kefniark/voice-your-choice

openhackday prototype speech-to-text text-to-speech voice-recognition webapi

Last synced: 7 days ago
JSON representation

Small Openhack prototype about Web Speech API

• Host: GitHub
• URL: https://github.com/kefniark/voice-your-choice
• Owner: kefniark
• Created: 2020-05-29T08:46:43.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2021-08-12T12:36:37.000Z (over 3 years ago)
• Last Synced: 2024-04-14T12:46:14.793Z (7 months ago)
• Topics: openhackday, prototype, speech-to-text, text-to-speech, voice-recognition, webapi
• Language: TypeScript
• Homepage: https://kefniark.github.io/voice-your-choice/assets/
• Size: 44.9 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 2
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Voice your Choice

## Description

Small Open Hack Prototype to experiment the current state of Web API about speech recognition.

[Live Demo](https://kefniark.github.io/voice-your-choice/assets/)

![Logo](./logo.png)

## Vocal Commands

* `Start` or `Next` (to change page)

* `Up`, `Down`, `Left`, `Right` or if you prefer `North`, `South`, `East`, `West` (it's the same)

## Conclusion

I managed to get something working but there are still lot of issues with those APIs:

* The quality of recognition doesnt seem to be super consistent

    * Score change quite a lot with just background noise

* Really slow, it takes between 1-3s to get the final result

    * It's possible to work with intermediate results, faster but they are even more inacurate :D

    * Seen some C library compiled to WASM on github, but haven't tested yet

* Browser implementation are far from good (example chrome ignore the `grammar` property)

I think their status "Experimental" is well deserved, and I don't recommend using them for real projects in their current state.

## Resources / Inspiration

* [Mozilla : Using the Web Speech API](https://developer.mozilla.org/en-US/docs/Web/API/Web_Speech_API/Using_the_Web_Speech_API)

* [Speech color changer](https://mdn.github.io/web-speech-api/speech-color-changer/)

## Usage

### Dev

```sh

npm install

npm run dev

```

### Build

```sh

npm run build

```