Stenotype Court Reporting (Ireland Stenotype, 1911)
Measured by National Court Reporters Association · Registered Professional Reporter certification standard (2024)
Inputs
The measured or assumed values behind the calculations, each with its source.
- rate = 225 wpm
- RPR certification Testimony/Q&A leg: 225 wpm. The higher RMR certification requires up to 260 wpm. This is the throughput benchmark; the system date comes from Stenograph's 1911 Ireland Stenotype history.
- P = 0.95
- 95% accuracy required to pass each leg; working reporters typically exceed this.
- H = 1.0 bits/char
- English-text entropy (Shannon).
- N = 30
- Output alphabet size for the raw-character Wolpaw ceiling (uniform prior over English characters). Not the per-stroke chord space, which is far larger; the transmitted information is bounded by the English text produced.
- T_char = 0.05333 s/char
- Gross character interval for the Wolpaw ceiling: 60 / (225 wpm × 5) = 0.05333 s. Gross because Wolpaw's accuracy term P handles errors separately.
Strictest ITR
Each scoring method is an upper bound on the channel, so the headline is the strictest (smallest) one for this entry. Use the score selector on the home page to view any single method across entries.
-
Characters per minute
225 wpm × 5 chars/word = 1125 chars/min
-
Net of errors
1125 × 0.95 = 1069 correct chars/min
Using the 95% certification accuracy threshold.
-
Information transfer rate
1069 char/min × 1.0 bit/char ÷ 60 s/min = 17.8 bits/s
What counts as a bit depends on the action space. The number of distinguishable actions and how likely each one is are design choices of the task, not the sensing hardware. The same modality can present a fixed set of targets, a set pruned per step by a grammar or language model, or a continuous control space. Each of these changes how many actions are live and how the probability mass is spread, and therefore the information per selection. Read the action space below before comparing headline numbers across entries.
Action space
What the user can produce at each step, and how those options are distributed.
- Structure
- Context-dependent (the live set changes per step)
- Size
- 30 distinguishable actions
- Prior
- Context-conditioned: likelihoods depend on prior actions
- Notes
- A chorded phonetic keyboard: each stroke can capture a whole syllable or word, and a steno theory/dictionary maps strokes to English. The reference counts the resulting English character stream at Shannon entropy. The per-stroke action space (thousands of chords) is much larger than the alphabet, but the transmitted information is bounded by the English text produced.
Comparability The strictest bound here is the Shannon entropy of the output text, under one predictor held constant across the whole atlas (≈1 bit per character). That shared predictor makes it directly comparable to every other text entry (keyboards, spellers, silent speech and speech BCIs) regardless of prior or vocabulary size. For most text interfaces it comes out tighter than the raw-selection bounds, but not always. Where a small vocabulary makes Wolpaw tighter, that wins instead. Any Fitts, Wolpaw or log₂(N) figure shown below is another bound on the same channel. Switch the home-page score selector to compare one across entries.
Other bounds considered for the headline
Also valid upper bounds for this entry and eligible to be the headline. They just came out looser than the strictest above. Pick any of these in the home-page score selector.
-
Bits per selection (Wolpaw formula)
B = log2(N) + P*log2(P) + (1-P)*log2((1-P)/(N-1)) = log2(30) + 0.95*log2(0.95) + 0.05*log2(0.05/29) = 4.378 bits / selection
Term 1 is the information if every choice were correct; terms 2-3 subtract the bits lost to the error rate, assumed spread evenly over the other N-1 targets.
-
Selections per second
T = 0.05333 s/selection -> 1 / 0.05333 = 18.751 selections/s
-
Information transfer rate
ITR = B * selections/s = 4.378 * 18.751 = 82.085 bits/s
Other score types
Bounds the atlas keeps out of the default strictest headline: as-reported figures, alternate task conditions, or raw-channel ceilings that shouldn't win the headline by default. Each still carries a score type, so the home-page selector ranks this entry on it when you choose that type. Read its derivation before comparing across entries.
-
Achieved-bitrate credit per net-correct character
N = 30 → log2(N − 1) = log2(29) = 4.86 bits per net-correct selection (field-standard achieved bitrate, e.g. Webgrid; Nuyujukian 2015, which introduced the metric, used log2(N)).
-
Net-correct character rate
net-correct = 2P − 1 = 2(0.95) − 1 = 0.90 of characters. At 0.05333 s/char (225 wpm) → 0.90 / 0.05333 = 16.88 correct/s.
A transcription error commits the wrong character rather than timing out, so incorrect = 1 − P. Same N (30), certification accuracy (95%) and character interval (0.05333 s) as the entry's raw-character Wolpaw ceiling, and lands on the same ~82 bits/s. As the entry notes, this per-character key channel loosely over-bounds steno, whose strokes capture whole syllables; the transmitted information is bounded by the 17.8 bits/s Shannon headline on the English text produced.
-
Achieved bitrate
4.86 bits × 16.88 correct/s = 82.0 bits/s.
Source
- Authors
- National Court Reporters Association
- Publication
- Registered Professional Reporter certification standard, 2024