https://github.com/membraneframework/membrane_whisper_plugin
https://github.com/membraneframework/membrane_whisper_plugin
Last synced: 2 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/membraneframework/membrane_whisper_plugin
- Owner: membraneframework
- License: apache-2.0
- Created: 2026-01-08T13:02:07.000Z (5 months ago)
- Default Branch: master
- Last Pushed: 2026-01-29T11:19:41.000Z (4 months ago)
- Last Synced: 2026-01-30T01:40:14.611Z (4 months ago)
- Language: Elixir
- Size: 10.2 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Membrane Whisper Plugin
[](https://hex.pm/packages/membrane_whisper_plugin)
[](https://hexdocs.pm/membrane_whisper_plugin)
[](https://circleci.com/gh/membraneframework/membrane_whisper_plugin)
A Membrane plugin for producing transcripts from audio streams containing human speech, based on OpenAI's Whisper model.
It's a part of the [Membrane Framework](https://membrane.stream).
## Installation
The package can be installed by adding `membrane_whisper_plugin` to your list of dependencies in `mix.exs`:
```elixir
def deps do
[
{:membrane_whisper_plugin, "~> 0.1.0"}
]
end
```
## Examples
For a demo streaming and transcribing a static `.mp4`, see `examples/live_mp4_processing.exs`:
```sh
$ elixir examples/live_mp4_processing.exs
```
You can also try it out yourself - see `examples/live_mic_processing.exs` which uses `portaudio` to stream your microphone's input directly to Whisper and prints the resulting transcript directly to console:
```sh
$ elixir examples/live_mic_processing.exs
```
## Copyright and License
Copyright 2026, [Software Mansion](https://swmansion.com/?utm_source=git&utm_medium=readme&utm_campaign=membrane_whisper_plugin)
[](https://swmansion.com/?utm_source=git&utm_medium=readme&utm_campaign=membrane_whisper_plugin)
Licensed under the [Apache License, Version 2.0](LICENSE)