30sec
🔍
Analyzing your request·
Building your personalized AI workflow
Pro tip
“The best workflows use 3-5 focused tools, not 10 mediocre ones.”
How the tools connect
Best tools for this job
Curated for this specific workflow - not a generic list
Step-by-step workflow
Follow this order for best results
1
Clean up the raw recording
Run your raw audio through speech enhancement to remove noise, echo and room sound before you start editing.
Why these tools
Cleaning first makes transcripts more accurate and the edit easier to judge.
Expected output
A clean, studio-like version of your raw recording.
2
Edit the content
Import the audio into a text-based editor, cut mistakes and tangents, remove fillers, and rearrange sections.
Why these tools
Descript gives you precise control while Cleanvoice batch-removes small verbal clutter automatically.
Expected output
A tightened episode with the story flowing well and distractions removed.
3
Polish the final sound
Export the edit and run a final enhancement pass if needed, then check levels so voices are consistent in volume.
Why these tools
A last pass evens out any differences created by cutting between clips.
Expected output
A consistent, listenable master audio file.
4
Add intro, outro and final review
Add your intro music and outro in the editor, then listen to the whole episode once at normal speed to catch errors.
Why these tools
You can finish everything in the same project without exporting to another app.
Expected output
A ready-to-publish MP3 or WAV file.
5
Create show notes and promo content
Upload the final episode to generate a transcript, show notes, title ideas and social posts, then upload to your podcast host.
Why these tools
Automates the time-consuming post-production writing so you can publish consistently.
Expected output
Show notes, transcript and social posts ready to publish.
Complete Guide
Beginner1 to 2 hours for a 30 to 45 minute episode once you are familiar with the toolsThis guide takes you from a messy raw recording to a polished episode with show notes. You do not need audio engineering experience: the tools do the technical work, and you make the creative decisions about what stays in.
Before you start
- ✓A raw audio or video recording of your episode
- ✓A free account on Adobe Podcast and Descript
- ✓Headphones for reviewing the edit
- ✓Optional: intro and outro music you have the rights to use
1
Clean up the raw recording
Open Adobe Podcast, go to Enhance Speech and upload your file. Listen to the result and compare with the original. If you recorded separate tracks for each speaker, enhance each one separately for the best results.
Example prompt
“Action: Upload raw-episode-12.wav to Enhance Speech, set strength to about 80 percent, and download the cleaned file.”
Pro tip
Do the cleaning before editing so the transcript is more accurate and you can hear problems clearly.
Common mistake
Pushing enhancement to the maximum, which can make voices sound robotic or watery.
2
Edit the content
Create a Descript project and import the cleaned audio. Wait for transcription, then read through and delete the parts you do not need. Use the filler word tool, or process the file through Cleanvoice first for a heavy cleanup.
Example prompt
“Action: In Descript, use Remove filler words, select 'um' and 'uh', review the list, and apply. Then delete off-topic sentences directly in the transcript.”
Pro tip
Keep some natural pauses and a few fillers so the conversation does not sound overly chopped.
Common mistake
Removing every pause and filler, which creates an unnatural, rushed feel.
3
Polish the final sound
Listen for jumps in volume or tone between clips after your cuts. Use Descript's studio sound or a quick extra pass in Adobe Podcast, and make sure all speakers are at similar loudness.
Example prompt
“Action: Apply Studio Sound at a low setting to the whole composition, then play back the sections where you made the biggest cuts.”
Pro tip
Aim for speech that is comfortable on both earbuds and car speakers.
Common mistake
Applying heavy processing twice, which stacks artifacts.
4
Add intro, outro and final review
Add intro and outro music on separate tracks, lower music volume under speech, and add short fades. Then listen to the full episode once from start to end before exporting.
Example prompt
“Action: Drop the intro file at the start, set music volume to about 20 percent under your voice, add a 2 second fade-out, and export as MP3 at 128 kbps mono or stereo.”
Pro tip
Listen on a different device than your editing headphones to catch problems.
Common mistake
Skipping the full listen-through and publishing with an obvious glitch.
5
Create show notes and promo content
Upload the final audio to Castmagic and generate the transcript, summary, show notes, and social post ideas. Read and edit the output to match your voice, then upload the audio and notes to your podcast host.
Example prompt
“Write show notes for this episode in a friendly tone: a 2 sentence summary, 5 key takeaways, timestamps for main topics, and 3 social posts for LinkedIn and X.”
Pro tip
Save your best show notes as a template so future episodes stay consistent.
Common mistake
Publishing AI-generated notes without checking names, links and facts.
Pro tips from power users
- 01Record in a quiet room with a decent microphone placed close to your mouth, since this improves every AI tool's results.
- 02Record each speaker on a separate track if you can for cleaner editing.
- 03Make a checklist of your steps so each episode takes less time.
- 04Keep your original raw files backed up until the episode is published.
- 05Use the transcript as a base for blog posts and newsletters.
Frequently asked questions
Can I do this entirely for free?
Mostly. Adobe Podcast and Descript have free tiers with limits on hours and file size, which can work for a few episodes a month. Heavier use will likely need a paid plan.
Do I need both Descript and Cleanvoice?
No. Descript can remove fillers on its own. Cleanvoice is optional if you want a more thorough automatic pass, especially for long episodes with many mouth sounds.
Will AI enhancement make my voice sound fake?
It can at high strength. Use a moderate setting and compare with the original so your voice still sounds natural.
What file format should I record and export?
Record in WAV if possible for best quality, and export the final episode as MP3 around 128 kbps, which suits most podcast hosts.
Can this work for video podcasts?
Yes. Descript edits video and audio together, so cuts in the transcript apply to both. You can still run the audio through Adobe Podcast first.
Next steps
How to move forward once your workflow is running
- →Track downloads, listen-through rate and listener feedback for your next five episodes to see what improves.
- →Create a repeatable template in Descript with your intro, outro and settings so editing is faster.
- →Use the transcript and Castmagic output to publish a blog post or newsletter for each episode.
- →Learn the basics of microphone placement and room treatment, since better source audio reduces editing time further.
Alternative paths
Different approaches depending on your priority
Explore more workflows
What else do you want to build with AI?
Ready to start this workflow?
Explore each tool or build a new workflow for your task