Character & Word Counter

The Character & Word Counter reads an SRT file and reports four numbers: the total number of cues, the total characters excluding all whitespace, the total words, and the average characters per cue to one decimal place. You paste the file into the SRT Content box and run it; the result is a short plain-text report rather than a modified subtitle file, and it downloads as a .txt.

Those figures drive real decisions. Translators and localisation vendors quote and invoice on character counts, and subtitle work is frequently priced per character or per word rather than per minute. Broadcasters and streaming platforms specify characters-per-line and reading-rate limits, and an average that has crept up over a series is an early warning that cues are getting too dense to read comfortably. Accessibility teams use word totals to sanity-check that a caption file actually covers the whole programme rather than stopping halfway. Getting these numbers by pasting subtitles into a word processor is unreliable, because it counts the timecodes and cue numbers as well.

The tool avoids that problem by parsing the file into cues first and measuring only the dialogue. Characters are counted after removing every whitespace character, which is the convention behind most subtitle character limits and translator rate cards, so spaces, tabs and line breaks inside a cue do not inflate the figure. Words are counted by splitting each cue's text on runs of whitespace and discarding empty results, so trailing spaces and blank lines do not create phantom words. The average is characters divided by cue count, and it safely reports zero rather than failing if the file contains no cues.

Nothing is modified and nothing is uploaded - your file is measured in the browser and the original text is left exactly as you pasted it.

• Four figures at once: A single run reports cue count, characters excluding spaces, total words and average characters per cue, which is usually everything a quote or a QC note needs.
• Dialogue only: Because the file is parsed into cues before measuring, timecodes and index numbers are excluded from the totals in a way a word processor cannot manage.
• Whitespace-free character count: All whitespace is stripped before counting, matching the standard used by subtitle character limits and most translator rate cards.
• Robust word splitting: Text is split on runs of whitespace with empty results discarded, so multi-line cues and stray trailing spaces do not distort the word total.
• Density indicator: The average characters per cue, given to one decimal place, is a quick proxy for whether your cues are getting too dense for comfortable reading.
• Read-only by design: The tool never rewrites your subtitles; the input you paste into the SRT Content box is returned to you untouched.
• Report you can keep: The output downloads as a .txt file, which is convenient to attach to an invoice, a purchase order or a QC record.
• Handles whole series files: Long concatenated tracks are counted in one pass with no cue limit.

1. Paste the subtitle file you want to measure into the SRT Content box, including cue numbers and timecodes - the tool strips them out for you.
2. Click Load sample first if you would like to see the shape of the report before running it on a real job.
3. Run the counter; there are no options, because the character convention it uses is the standard one for subtitle pricing and line limits.
4. Read the four lines of output: cues, characters excluding spaces, words, and average characters per cue.
5. Compare the cue count against what you expect from the programme length, since a suspiciously low number usually means you pasted a partial file.
6. Download the report as a .txt file and attach it to your quote, invoice or QC record so the figures are documented alongside the delivery.

Example 1 - quoting a translation job. A short promo file is measured before pricing.

Input (extract)
1
00:00:01,000 --> 00:00:03,000
Welcome to the summer season.

2
00:00:04,000 --> 00:00:06,500
Tickets are on sale now.

Report
Cues: 2
Characters (no spaces): 47
Words: 9
Avg chars/cue: 23.5

Note that the 47 excludes every space and covers only the dialogue, not the timecode lines.

Example 2 - a density check across an episode. A 640-cue episode reports an average of 71.2 characters per cue. Against a two-line, 42-characters-per-line house limit that is uncomfortably close to full, telling the editor to split the longest cues before the reading-rate check rather than after it.

• Quote and invoice with confidence: You have a defensible character and word figure for the actual dialogue, not a word-processor count polluted by timecodes.
• Catch density problems before QC does: A rising average characters-per-cue flags reading-rate trouble while there is still time to split the cues.
• Verify a file is complete: A cue count that does not match the programme length is an immediate signal that something was truncated in transfer.
• Zero risk to the asset: The tool only reports, so you can measure a locked delivery file without any chance of altering it.
• Documentation for the paper trail: The downloadable .txt report can be filed with the job, which settles disputes about scope after the fact.
• Client material stays private: Counting happens in your browser, so you can measure confidential scripts without uploading them anywhere.