How to Specify Duration Accurately When Ordering Narration: Characters per Minute, Script Adjustment, and Time-Stretch Limits

How to Communicate Duration Requirements Accurately When Ordering Narration

One of the most common problems in narration projects is a mismatch in duration expectations.
A client may say “make it one minute,” but for the narrator, that could mean either “about one minute in a natural read” or “exactly 59.5 seconds, locked to picture and music cues.” That difference changes how the session is approached. If this is unclear, you often end up with “a little too long” or “too short” after recording, which quickly increases revision costs.

The core rule is simple: always communicate these three items together:

1. Target duration: for example, “exactly 60 seconds,” “58–60 seconds,” or “15-second commercial length”
2. Tolerance: for example, “within ±0.5 seconds” or “a little variation is acceptable”
3. Priority: for example, “natural delivery first” or “timing accuracy first”

With just these three points, decision-making on the recording side becomes much more accurate.

How Many Characters Fit in One Minute?

A very common question is: “How many Japanese characters fit into one minute?”
In practice, the answer depends heavily on the type of narration. As a working benchmark:

• Slow, emotional delivery: 180–220 characters per minute
• Standard corporate or web video: 230–280 characters per minute
• Dense e-learning or explainer content: 280–330 characters per minute
• Fast promotional or infomercial style: 330–380 characters per minute

The key point is that character count alone is not enough.
Even with the same 250 characters, a copy-driven line like “Today, the world changes.” takes different time than a technical script filled with product names and numbers.

Factors that often make a script run long include:

• Heavy kanji usage that is harder to speak than to read
• Many numbers, units, English words, or model names
• Frequent proper nouns or technical terms
• Few commas, making breath points unclear
• Emotional delivery or dramatic pauses

So when ordering narration, do not share only the character count. Share the type of script as well.
For example: “265 characters, explanatory narration with many product names and model numbers,” or “210 characters, brand film copy that relies on pauses.” That improves timing estimates dramatically.

How to Adjust a Script That Does Not Fit the Duration

In many cases, duration problems come from script design, not recording skill.
Before forcing the narrator into an unnaturally fast read, fix the script first. My practical adjustment order is:

1. Remove words that do not change meaning

Examples in Japanese often include trimming polite redundancy or filler phrasing.
This alone can reduce total length by 5–10%.

2. Merge overlapping information

A sentence that repeats similar ideas can often be compressed into one cleaner phrase without losing meaning.

3. Rewrite hard-to-read structures

When time is tight, awkward sentences work against you. They increase stumbling risk and often make the read longer.
Clear punctuation, better word order, and one message per sentence usually make the delivery both faster and more natural.

4. Rank what is essential

Mark the script in three layers:

• Absolutely required
• Preferably kept
• Can be removed

If you share this in color coding or comments, the narrator and director can make better timing decisions.

A highly practical method is to prepare not just one final script, but two shorter backup versions as well. For example:

• Version A: full script, 290 characters
• Version B: 10-second shorter version, 255 characters
• Version C: minimum essential version, 230 characters

This reduces retakes and makes editing much easier.

How Far Can You Rely on Time-Stretching After Recording?

Clients sometimes ask: “If it runs a little long, can’t we just shorten it later?”
The answer is: time-stretching is a rescue tool, not a universal solution.

Tools such as Adobe Audition, iZotope RX, Pro Tools, and Logic Pro offer high-quality time-stretch functions.
However, the natural-sounding limit is usually around:

• ±2–3%: generally safe in practice
• Around ±5%: usable with caution
• Above ±8%: noticeable artifacts become likely
• Over ±10%: re-recording is usually the better choice

When shortening audio, consonant attacks and line endings can start to sound unnatural.
When lengthening, you may hear dragging, unnatural resonance, or formant issues.
This is especially obvious in brand films, quiet documentaries, or any project where pauses themselves are part of the performance.

On the other hand, in e-learning or internal training audio—where information delivery is the main priority—adjustments of around 2–4% can sometimes work well.
So the real question is not “Is it technically possible?” but “Does it still sound natural in the finished piece?”

A Duration-Spec Template You Can Use Right Away

Here is a practical format you can paste into your narration request:

• Target duration: exactly 60 seconds
• Tolerance: ±0.5 seconds
• Priority: timing accuracy over natural looseness
• Script length: 248 characters
• Script characteristics: few numbers or English words, corporate introduction, standard tempo assumed
• Performance request: slower opening, slightly tighter pace in the second half
• Backup plan: if it exceeds 60 seconds, please also record shortened Version B

When you specify requests at this level, the narrator can design where to pause and where to tighten pace.
A duration spec is not just a number in seconds. It is the combination of target time, tolerance, priority, and script characteristics.
Once that is clear, the quality of the first take improves dramatically. In narration work, the precision of the brief directly shapes the quality of the voice track.