Common-voice: Download personal recorded dataset

Created on 3 Jul 2018 · 8Comments · Source: mozilla/common-voice

One of the local researchers is trying to collect the voice samples of ALS patients, in order to synthesize the patient's voice when necessary. I feel we can really help with a "download personal dataset" feature.

Enhancement help wanted

Source

irvin

👍8

All 8 comments

This usecase is also good for building TTS (speech synthesis) based on your own voice.

mikehenrty on 3 Jul 2018

And I feel it’s also something we should implement in order to respect the principle of GDPR data portablility right.

irvin on 3 Jul 2018

🚀1

This use case can also be of interest in creating a STT engine for, say, people with a particular speech impediment.

kdavis-mozilla on 7 Jul 2018

As Irvin mentions, we will have to do this due to GDPR. Plus it's a nice feature to have.

mikehenrty on 9 Jul 2018

434 is a related issue (maybe we should merge them)

Gregoor on 9 Jul 2018

Settings Build Profile_06 1

We have designs for it, but so far haven't found time to look at implementation. Happy to review PRs for it!

Gregoor on 22 Jul 2019

❤1

I'd like to start working on this feature. This will be my first contribution, so let me know if you have any advice or additional information I should know.

Thanks!

kjreills on 1 Feb 2020

👍1

@kjreills Please feel free! If you run into any issues getting the repo set up locally for development please submit a PR to CONTRIBUTING.md so that we can improve our documentation :)