Adobe Podcast
Free tierAI-powered audio recording and editing, all on the web
Key strengths
Technical Integration & Usage
Enhance Speech (Standalone Tool)
The Enhance Speech endpoint is accessible via the web UI and (in limited beta) via API. Upload audio/video files up to a set size limit; the model runs server-side inference and returns a cleaned audio file. Supported input formats include MP3, WAV, M4A, and MP4.
Browser Recording Architecture
Adobe Podcast uses WebRTC for real-time multi-participant recording. Each speaker's audio track is captured and stored independently (multi-track), enabling per-speaker enhancement and editing post-session.
Transcript-Based Editing
Transcription is powered by Adobe Sensei's ASR (Automatic Speech Recognition) pipeline. The transcript is word-level time-aligned with the audio waveform, so deleting text nodes in the editor programmatically trims the corresponding audio segment.
Creative Cloud Integration
Assets recorded or enhanced in Adobe Podcast can be pushed to Adobe Premiere Pro and Adobe Audition via the Creative Cloud Libraries API, preserving metadata and multi-track structure for downstream editing workflows.
