What a standard would have to say: forty-seven reporting clauses for acoustic voice measurement, each with the measurement that justifies it

In 1904 Baltimore burned for thirty hours while fire engines from Washington, Philadelphia and New York stood in its streets unable to help, because their hose couplings did not fit the city's hydrants. The threads were not a hard engineering problem. They were an agreement nobody had made, and the cost of not having made it became visible only when the city was on fire. Acoustic voice measurement has no such agreement. This record proposes one: forty-seven reporting clauses, each stated with the measurement that justifies it, condensed into a fourteen-line checklist that any paper could fill in. Thirty-nine clauses come from a book-length audit of the field; eight are new and follow from seven open records deposited alongside this one. THE CLAUSES ARE RE-EVIDENCED, not repeated. Six carried numbers derived from a single vowel at a single pitch; those have been recomputed across five vowel and pitch conditions on 8,191 recordings with an electroglottographic truth, and reported with confidence intervals. The response slope of jitter, previously stated as 0.34, is 0.296 (95 % CI 0.270-0.324). Its collapse in clean recordings, previously stated as reaching 0.04, reaches 0.012 (0.002-0.028) -- and is now shown not to be range restriction. THE DESIGN PRINCIPLE. Forty-five of the forty-seven clauses say REPORT X. Two say USE THE VALUE Y, and both are marked. That distinction answers the only serious objection to a standard arriving this early: a reporting requirement freezes nothing. It makes current practice visible, which is the precondition for improving on it. If everybody reports their harmonic bandwidth, the field can then discover which choice is better, and the discovery becomes possible precisely because the parameter has been brought into the open. The two prescriptive clauses are the two with a direct measurement behind them, stated so that anybody who disagrees knows exactly what they must overturn. TWO OF OUR OWN PUBLISHED FIGURES ARE CORRECTED. The gap between clinical practice and the thermodynamic noise floor of air was described in this project's book proposal as forty-three decibels; 43.5 dB is the gap from a sound-treated booth, and clinical practice sits 70.4 dB above the floor. And the specification was described as fifty-five clauses; the first edition contains thirty-nine. NOTHING HERE IS EXPENSIVE. Twelve of the fourteen lines are numbers the analysis has already computed and simply not printed. The reason they are absent is not cost. It is habit. The fourteen-line form is deposited separately and may be reproduced freely, in whole or in part, with or without attribution.