No project description provided
Project description
Whisper-dictation
This is an app using openai's whisper to dictate on KDE wayland.
The project is designed as a dictation server that runs at background (To avoid the time to load model each time starts the dictation) and a client to toggle if the server should be recording. You can assign a shortcut to toggle the server to start/stop the recording (whisper_dicatation --language [language_code]).
Whenever the dictation is stopped, the content will be sent to your clipboard and a notification will be displayed.
The project depends on the kdialog package and wl-copy (from wl-clipboard package).
This project is designed to work on KDE wayland. Other wayland platforms might work as well, but without the ability to send a notification.
Installation
Install using pip:
pip install whisper_dictation
Note that it is recommended to install at user directory (not globally). Since the systemd service provided is written only for executable at ~/.local/bin
Usage
To start the project manually, you should use two terminals, for the server:
whisper_dictation daemon [--port 9000] [--model_name base]
See whisper_dictation daemon --help for all the available models
You can use a command to trigger the daemon, or assign a shortcut to this command in order to use it. Press once for start, and press the second time to stop the recording.
whisper_dictation say [--language en]
You should assign a language code, it can help with the performance especially using a small model.
Alternatively, you can use the systemd service unit provided inside this repo to make the daemon running in the background. Place it in your ~/.config/systemd/user/, enable and start it:
systemctl --user enable whisper_dictation
systemctl --user start whisper_dictation
TODO
- add system integration for a shortcut to start/stop dictation
- output the dictation to where the cursor is (planned as fcitx addon).
- optional(A system tray)
- package it on aur
Requirements
wl-clipboardkdialog
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file whisper_dictation-0.4.3.tar.gz.
File metadata
- Download URL: whisper_dictation-0.4.3.tar.gz
- Upload date:
- Size: 17.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/1.4.2 CPython/3.10.10 Linux/6.2.10-zen1-1-zen
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
98f524badf3c0aebea7fd8eeeaffa32dd31700640eebae01c23fe091e0c2d51a
|
|
| MD5 |
0cad3842d6475ae045a2df85a94a4f8c
|
|
| BLAKE2b-256 |
811fc1cb9f99c032104fe8db24893ae76ee5dcd8613ae275cbf4c9b862851611
|
File details
Details for the file whisper_dictation-0.4.3-py3-none-any.whl.
File metadata
- Download URL: whisper_dictation-0.4.3-py3-none-any.whl
- Upload date:
- Size: 19.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/1.4.2 CPython/3.10.10 Linux/6.2.10-zen1-1-zen
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
85ff214cca3ae652e84fe72ace894cd5397077df80eb036b049a32b72c99e292
|
|
| MD5 |
8efac0faf898ad6c1a8f755dde03f8f4
|
|
| BLAKE2b-256 |
5ab84a2be8c286957751e21fcd274547e6707e5963d38e6cf75293f09f9e34fa
|