1
0
Fork
You've already forked dispatch-caption
0
Speech to text project, aimed at providing real-time, good-enough, text output from dispatch radio
  • Go 100%
2024年11月03日 07:16:06 -05:00
.gitignore begin restructure to set audio source from CLI flag 2024年09月14日 13:10:54 -04:00
fdnyems-7-1728777877.m4a sample file as retrieved from openMHz (no reencode) 2024年10月14日 12:18:21 -04:00
fdnyems.wav representative audio sample for testing 2024年09月14日 12:38:04 -04:00
go.mod update deps 2024年10月31日 15:43:34 -04:00
go.sum update deps 2024年10月31日 15:43:34 -04:00
LICENSE Initial commit 2024年09月12日 14:35:07 +00:00
main.go f typo 2024年11月02日 09:08:18 -04:00
README.md fix list nesting in README and add comment on package managers, portaudio 2024年11月03日 07:16:06 -05:00

dispatch-caption

Speech to text project, motivated by a desire for real-time, good-enough, text output from dispatch radio

Build and Install

  1. If you have a system package manager, most of the requirements should be available there and that is the recommended way to install and manage the following components.

  2. A Go (golang) development environment: see https://go.dev/doc/install

  3. Setup Google Cloud credentials to run:

     gcloud init
 gcloud auth application-default login

  4. Download and install PortAudio development library: https://files.portaudio.com/download.html

  5. Clone repository and build with go build .

  6. Run binary locally:

    • ./dispatch-caption -s to capture from the microphone
    • ./dispatch-caption --file fdnyems-7-1728891544.m4a to test a file sample
    • ./dispatch-caption --help to see all options

Notes

Sample audio files retrieved from OpenMHz