Famunkwriter
Posts: 1
Joined: Mon Feb 24, 2020 12:53 am

Captain’s log voice and transcription

Mon Feb 24, 2020 12:59 am

Have an idea for a project and am trying to find out if this is even possible.

What i want to do is use a Raspberry Pi to record a voice memo by pushing a button or flicking a switch. The maximum recording time would be around 5-10 minutes. The recording would stop when the button is pushed again, and the file would then be transcribed and saved in a text file with the current date and timestamp as the filename to a folder on a specified hard drive.

Ive read that the transcription bit will typically require an internet connection but thought I’d see if the idea is even achievable without it.

Thanks for your help!

User avatar
topguy
Posts: 6527
Joined: Tue Oct 09, 2012 11:46 am
Location: Trondheim, Norway

Re: Captain’s log voice and transcription

Mon Feb 24, 2020 2:56 pm

After reading this thread:
https://unix.stackexchange.com/question ... -for-linux

Only https://github.com/mozilla/DeepSpeech caught my eye..

..can it run on a Pi ?
Quote from https://hacks.mozilla.org/2019/12/deeps ... xt-engine/
DeepSpeech v0.6 with TensorFlow Lite runs faster than real time on a single core of a Raspberry Pi 4.

User avatar
neilgl
Posts: 2226
Joined: Sun Jan 26, 2014 8:36 pm
Location: Near Aston Martin factory

Re: Captain’s log voice and transcription

Mon Feb 24, 2020 8:59 pm

To record in a file (a.wav) using a USB microphone:

Code: Select all

arecord -D sysdefault:CARD=1 -d 10 -f cd -t wav a.wav

Return to “Beginners”