Episode Lifecycle

Every episode in EnaCast goes through a series of automatic stages from the moment it’s recorded until listeners can hear it. The process is fully automatic, although editors can step in at any point if needed.

Overview

  ┌─────────────┐
  │  Recording   │  Audio is automatically captured from the stream
  └──────┬───────┘
         │
  ┌──────▼───────┐
  │  Processing  │  Uploaded to the cloud, waveform generated
  └──────┬───────┘
         │
  ┌──────▼───────┐
  │Transcription │  Audio is converted to text (if enabled)
  └──────┬───────┘
         │
  ┌──────▼───────┐
  │ AI Analysis  │  Title, summary, tags, and sections generated
  └──────┬───────┘
         │
  ┌──────▼───────┐
  │ Trimming     │  Silences and pre-show content removed
  │(auto/manual) │
  └──────┬───────┘
         │
  ┌──────▼───────┐
  │ Publication  │  Available on the web and RSS feed
  └──────────────┘

Stages in Detail

1. Recording

When airtime arrives according to the show’s configured schedule, EnaCast automatically starts recording audio from the stream. Recording stops when the scheduled time ends.

Note

Recording only happens if the show has recording enabled and the station is not in holiday mode. Reruns (marked as such in the schedule) are not recorded.

The recorded audio is saved as an MP3 file and a new episode is created, linked to the corresponding show.

2. Processing

Once recording finishes, the episode is automatically processed:

• The audio file is uploaded to the cloud to ensure permanent availability
• The waveform (visual representation of the audio) is generated for the trim editor
• The episode’s duration is calculated

This happens in the background and usually takes just a few minutes.

3. Transcription

If the show has transcription enabled and daily budget is available, the audio is automatically transcribed. Transcription converts audio to text, which then serves as the basis for AI analysis.

The transcription level configured for the show determines how much audio is processed:

Level	What Is Transcribed
Full	The entire episode
15 minutes	Only the first 15 minutes
30 minutes	Only the first 30 minutes

For more details on levels and budget, see Transcriptions and AI Insights.

4. AI Analysis

From the transcription, artificial intelligence automatically generates:

• A descriptive title for the episode’s content
• A summary of the topics covered (long and short versions)
• Tags to categorize the episode
• Topic sections with automatic markers
• Highlighted quotes with informational value
• Start and end detection of the actual content

Tip

With partial transcription (15 or 30 minutes), only an automatic title is generated. For the full analysis, use full transcription.

If the AI detects where the actual program content starts and ends (for example, it detects that the first 2 minutes are a news bulletin), these trim points are saved for automatic or manual trimming.

5. Trimming

Trimming removes unwanted parts of the episode (silences, pre-show bulletins, filler music at the end). It can happen in two ways:

Automatic trimming: if the show has automatic trimming enabled, the AI detects the start and end points and trims the episode with no editor intervention needed. For more information, see the automatic trimming section in Transcriptions and AI Insights.

Manual trimming: the editor opens the trim editor from the admin panel and sets the desired start and end points. See Trim and Edit a Podcast for how to do this.

The original audio is always preserved

Trimming never modifies the original recording. Only a trimmed version is created. This allows re-trimming the episode if needed.

6. Publication

The episode is automatically published in two places:

• Station website: the episode appears on the show’s page with the audio player, transcription, section markers, and all AI-generated information.
• RSS feed: the episode is included in the show’s RSS feed, compatible with all podcasting platforms (Spotify, Apple Podcasts, etc.).

Wait times before RSS publication depend on the episode’s status:

Status	Wait Time
Trimmed (manually or automatically)	5 minutes
Not trimmed	30 minutes
Pending transcription or analysis	Until processing finishes

For more details on RSS distribution, see RSS Podcast Distribution.

What Can the Editor Do at Each Stage?

Stage	Editor Actions
Recording	No intervention needed. It’s automatic
Processing	Wait for it to finish. Usually a few minutes
Transcription	No intervention needed. If it fails or there’s no budget, the episode is published without transcription
AI Analysis	Review and modify the generated title, summary, and tags if desired
Trimming	Adjust trim points manually if automatic trimming wasn’t precise, or trim manually if automatic trimming isn’t enabled
Publication	Edit the episode’s public information: title, description, image

Visual Summary

Airtime
    │
    ├─ Is the station on holiday?
    │     YES → Nothing is recorded
    │     NO ↓
    │
    ├─ Does the show have recording enabled?
    │     NO → Not recorded
    │     YES ↓
    │
    ├─ Is it a rerun?
    │     YES → Not recorded
    │     NO ↓
    │
    ├─ Recorded automatically
    │
    ├─ Uploaded to cloud, waveform generated
    │
    ├─ Is transcription enabled with available budget?
    │     NO → Published with the show's default info
    │     YES ↓
    │
    ├─ Transcribed and analyzed by AI
    │
    ├─ Is automatic trimming enabled?
    │     NO → Editor can trim manually
    │     YES → Trimmed automatically
    │
    └─ Published on the web and RSS feed

Related Pages

• What Is EnaCast? — Introduction to the platform
• Transcriptions and AI Insights — Transcription levels, budget, and automatic trimming
• Trim and Edit a Podcast — How to manually trim an episode
• RSS Distribution — RSS feed publication timing
• Edit a Podcast — How to modify an episode’s public information

---

If you have questions, contact technical support at support@enacast.com