How to Use Descript to Edit Audio with AI

Descript transforms audio editing by letting you edit recordings like documents through AI-powered transcription. Its artificial intelligence features automate common tasks like removing filler words, creating voice clones, and cleaning up audio quality.

1. Import your audio file. Open Descript and click New Project. Drag your audio file directly into the workspace or click Add File > Audio. Descript automatically begins transcribing your audio using its AI engine, which typically takes 1-2 minutes per hour of audio.
2. Enable AI transcription corrections. Click the Settings gear icon in the transcription panel. Turn on Speaker Detection to automatically identify different speakers. Enable Punctuation & Capitalization for properly formatted text. These AI features improve transcription accuracy by 15-20%.
3. Remove filler words automatically. Select your entire transcription by pressing Cmd+A (Mac) or Ctrl+A (Windows). Go to Edit > Remove Filler Words. Choose your removal intensity: Light removes obvious filler words, Aggressive removes more hesitations and repetitions. Click Apply to let AI clean your audio.
4. Edit audio by editing text. Click anywhere in the transcription to position your playhead. Delete words by selecting text and pressing Delete—this removes the corresponding audio. Cut, copy, and paste sections of text to rearrange your audio. Add pauses by pressing Enter to create line breaks.
5. Create an AI voice clone. Select a section of clear speech that's at least 10 seconds long. Right-click and choose Create Voice Model. Name your voice model and click Create. Descript's AI analyzes speech patterns, tone, and pronunciation to build a synthetic version of the voice.
6. Generate speech with AI voice. Type new text directly into the transcript where you want to add speech. Highlight the new text and click the waveform icon that appears. Select your created voice model and click Generate. The AI creates audio that matches the original speaker's voice characteristics.
7. Apply AI audio enhancement. Select your audio track and open the Effects panel on the right. Enable Studio Sound to remove background noise and improve clarity automatically. Adjust EQ settings using the AI-powered presets like Podcast Voice or Interview Cleanup for instant audio improvement.
8. Export your edited audio. Click Publish in the top right corner. Choose Audio File as your export format. Select your desired quality settings—320kbps MP3 for web use or WAV for highest quality. Click Publish to render your AI-edited audio file.