Your 30-Second Safety Net
The audio buffer is the feature that makes 12scribe fundamentally different from a simple voice recorder. While you record, 12scribe keeps a rolling window of audio in memory — a continuously updating snapshot of the recent past. When you tap Add Marker, the marker does not start at the second you tapped. It automatically reaches back and captures the audio that happened before your tap.
Here is why this matters: human reaction time creates an unavoidable delay. A professor states a key definition. Your brain processes it, decides it is important, and your finger taps the screen. By the time all of that happens, three to ten seconds have already passed. Without a buffer, your marker would start after the definition is over — capturing silence or the next topic. With the buffer, the marker captures the complete definition, including the context leading up to it.
The default buffer is 30 seconds. That means every time you tap a marker, the recorded audio for that marker starts 30 seconds before your tap. React late, capture early. The buffer has you covered.
How the Buffer Works
The buffer runs continuously and silently from the moment you press Record. You do not need to enable it, arm it, or think about it. It is always there, always capturing, always ready.
When you tap Add Marker at, say, timestamp 02:15:
- With a 30-second buffer (default), the marker starts at 01:45
- With a 60-second buffer, the marker starts at 01:15
- With a 10-second buffer, the marker starts at 02:05
During playback, tapping that marker jumps you to its start time — the buffered position, not the moment you tapped. This means you hear the full context: the important thought from its beginning, not from the moment you reacted to it.
Duration Options: Which One to Choose
12scribe offers six buffer durations. Each serves a different recording style and content density. Here is a practical guide to choosing the right one:
- 10 seconds — best for rapid-fire content. Quick Q&A sessions, panel discussions with fast speaker changes, or situations where you are marking many moments close together. Short buffers keep marker segments compact and tightly focused, making review faster. Choose this when precision matters more than coverage.
- 20 seconds — a middle ground for moderately paced content. Team standups, structured meetings with clear agenda items, or casual conversations where the important point is usually a single sentence you want to capture.
- 30 seconds (default) — the balanced option that works for the majority of recordings. Lectures, interviews, one-on-one meetings. Thirty seconds is long enough to capture a typical important statement with full context, short enough to keep markers focused. Start here and adjust only if needed.
- 45 seconds — for content where ideas take longer to develop. Graduate-level lectures, technical presentations, or detailed explanations where the key insight builds over 30-40 seconds before reaching its conclusion. If you often find that 30 seconds clips the beginning of the thought you wanted, try 45.
- 60 seconds — extended coverage for dense academic content. University lectures where a single concept might take a full minute to explain, long-form interviews where the interviewee builds an argument over multiple sentences, or any situation where you tend to react slowly because you are multitasking. With a 60-second buffer, even if you space out for a full minute, one tap captures everything.
- 90 seconds — maximum coverage. For research recordings, doctoral seminars, or lectures where key arguments build over 60-90 seconds before reaching their conclusion. Use this when you want to guarantee that no amount of delayed reaction can cause a missed thought. The tradeoff is that each marker covers more audio, so review takes slightly longer.
Real-World Scenarios
These examples show how the buffer duration affects what gets captured:
- The late reaction — A professor starts defining "entropy" at 14:30 and finishes at 14:45. You realize it was important and tap at 14:50. With a 30-second buffer, the marker starts at 14:20 — the entire definition is captured, plus 10 seconds of context before it. You never miss a thought.
- The multitasker — You are taking handwritten notes during a meeting while recording. A colleague announces a budget change, but you are mid-sentence in your notes and do not tap until 20 seconds later. With a 30-second buffer, the announcement is fully captured. With a 10-second buffer, it might be partially clipped.
- The dense lecture — A graduate professor explains a theorem over 70 seconds. You tap at the end. With a 30-second buffer, you get the final 30 seconds. With a 90-second buffer, you get the entire explanation from start to finish — one tap, complete capture.
- The rapid marker — During a quick brainstorming session, you add a marker every 30-45 seconds. A 10-second buffer keeps each marker tight and focused, avoiding overlap with the previous one. A 90-second buffer would create significant overlap between consecutive markers.
Per-Project Settings
Buffer duration can be configured at two levels, giving you flexibility to match different recording contexts without constantly changing settings:
The global setting in the Settings tab sets the default buffer duration for all new projects. When you create a new project, it inherits this value automatically. Change the global setting to match your most common use case.
The per-project setting inside each project's configuration overrides the global default. This lets you use a 30-second buffer for weekly team meetings, a 60-second buffer for university lectures, and a 10-second buffer for quick voice memos — all configured independently.
To change the buffer for a specific project, open the project, navigate to project settings, and adjust the Buffer Duration slider. The change applies to all future recording sessions in that project. Existing recordings and their markers are not affected — the buffer only determines what is captured at the moment you tap a marker.