Just a heads up!
You’ll now see the Exact match filter before the Approximate match text box.
How to Use Cursecut
1. Upload Your File
Click the "Browse..." button to select a video or audio file. The maximum file size is 2GB per file. Supported formats include MP4, MOV, MP3, WAV and more. To increase accuracy, it is recommended to isolate voice tracks if possible and upload only those. If you have multiple files, you can use our new batch upload feature to process them all at once. Just select multiple files in the popup window when you click "Browse".
2. Select Audio Language
Choose the language of your audio from the dropdown menu. This helps improve the accuracy of word detection. If you're unsure or your content contains multiple languages, leave it blank for automatic detection. Our system supports over 30 languages and can automatically identify the primary language used in your audio. It is always recommended that you specify the main spoken language for accuracy reasons.
3. Set Words to Filter
Enter the words you want to censor in the "Words to filter" section. Separate multiple words with spaces. Choose between:
- Approximate match: Catches variations of words (e.g., plurals, tenses). For example, entering "curse" would catch "cursing" and "cursed".
- Exact match: Only censors the exact word entered. Use this for more precise control, especially for shorter words.
4. Precise Word Censoring
This feature allows for more targeted censoring of specific parts of words. It works in conjunction with the "Approximate match" option:
- Substring Detection: The system scans for the exact substring specified in the "Approximate match" field within larger words.
- Partial Censoring: Only the detected substring is censored, leaving the rest of the word intact.
- Position Sensitive: The censoring is applied only where the substring is found, maintaining the word's structure.
Example of how it works:
- Word to censor: "motherf**cker"
- Approximate match entered: "f**k"
- Result: "mother****er"
If you were to include the whole word "f**cker" in the Approximate match section it would censor to "mother******" instead.
This approach ensures minimal interference with the text while effectively targeting specific problematic substrings.
Note: This feature is currently available only for English language text.
5. Adjust Word Retention
Use the slider to set the percentage of each censored word to keep. This feature allows for nuanced censorship:
- 0%: Full censorship, the entire word is bleeped out.
- 50%: Half the word is kept, half is censored. For example, "f**king" might become "fu".
- 75%: Most of the word is kept, with just a small portion censored.
Experiment with different percentages to find the right balance between content integrity and appropriate censorship. Note that this will apply to every word, which is not ideal for every word. We are currently working on a solution to this.
6. Choose Bleep Method
Select how you want censored words to be handled. Options include:
- Silence: Replaces the word with complete silence. E.g., "What the [silence] is going on?"
- Sine Bleep: Uses a high-pitched beep. E.g., "What the [beeeeeep] is going on?"
- Custom sound (Beta): Upload your own censor sound (e.g., a duck quack). Click "Custom sound", then upload a short audio clip (.mp3 or .wav). This audio clip needs to be really concise and shouldn't have any silence in it for optimal effect.
- Reverse audio (Beta): Plays the censored word backwards. E.g., "What the [etitelp-xe] is going on?"
Beta features may have limitations. If you encounter issues, try the Silence or Sine Bleep options.
7. Process Your File
Click the "Process" button to start censoring your audio or video. The processing time depends on the file size and the number of words being filtered. For large files or batch processing, you may want to leave the tab open and let it run in the background. You'll see a progress bar indicating how far along the process is.
8. Review Output
Once processing is complete:
- Preview the censored video in the "Example Output" section.
- View the transcription with timestamps of censored words.
- Download the processed file for your use.
- If needed, adjust your settings and use the free re-processing feature to refine the results.
New Features
We've recently added:
- Batch upload: Process multiple files at once for increased efficiency.
- Customizable word length: Fine-tune your filtering for better accuracy.
- Free re-processing: Adjust your filter words or bleep method and process again at no extra cost.
- Improved accuracy: Added extra accuracy for languages like English.
- Smart precision filtering: For english content, we detect the keywords with more granularity so whole compound words are not completely censored.
Privacy Note
Your privacy is our top priority. All audio processing is done entirely on your device using advanced browser technologies. No audio or video data is ever stored on our servers. Transcription is done using our own custom model designed for this use case. Note that the transcription might make some errors and you should always review the output manually.