Skip to main content
Speech to text can be used to create a verbatim transcript of the audio; however, some content is not documented in the same manner as it is verbalized. The formatting features provide control over how key information should for represented in the textual output.

Feature availability:

/transcribe

/streams

/transcripts

Notes:
  • Transcript output is formatted according to the default options listed below for both transcribe and streams endpoints.
  • Add formatting configuration to transcribe connections to change output preferences - see details here.
  • Formatting configuration is not supported on streams connections.

Formatting Options

Use the following values within the formatting configuration object to define preferences:

The tables below show available options for each formatting configuration parameter:
  • The examples represent how output would be formatted given the spoken input shown for the as_dictated option.
  • The Default represents server default values, indicating output format when no value is defined for a given formatting parameter.
  • The Date and Time options indicated as locale will format according to local standards, dependent on the primaryLanguage defined in the configuration.

Dates

Date formatting allows for standardization of date output when dictating the year, month, and/or day. In the table below, the examples for each option indicate how the date would be formatted for spoken form showed in the as_dictated option.
The Date parameter options defined above are the go-forward standard; however, the following legacy date formatting parameter values remain supported by the API:

Times

Numbers

Notes:
  • Localization of numbers applied automatically for thousands and decimal separators.
  • numerals_above_nine applies to integers: When using this value, all whole numbers less than ten will be formatted as words, and all decimal numbers will be formatted as numerals. Use as_dictated for decimal numbers to be output as words instead of numerals.
  • A maximum of three decimal places is supported at this time. Be sure to dictate the numbers after decimal individually. For example, “one point two five” (not one point twenty-five) will return “1.25”.

Units and Measurements

  • Celsius (°C)
  • Fahrenheit (°F)
  • Kelvin (K)
  • Inch (in)
  • Meter (m)
  • Centimeter (cm)
  • Millimeter (mm)
  • Nanometer (nm)
  • Kilometer (km)
  • Square meters (m²)
  • Square centimeters (cm²)
  • Gram (g)
  • Kilogram (kg)
  • Milligram (mg)
  • Microgram (mcg)
  • Nanogram (ng)
  • Pound (lb)
  • Ounce (oz)
  • Liter (L)
  • Hectoliter (hL)
  • Deciliter (dL)
  • Centiliter (cL)
  • Milliliter (mL)
  • Picoliter (pL)
  • International unit (IU)
  • Milliequivalent (mEq)
  • Mole (mol)
  • Millimole (mmol)
  • Blood pressure (systole/diastole)
  • Beats per minute (BPM)
  • Respirations per minute (RPM)
  • Body mass index (BMI)
  • Percent (%)
  • Millimeters of mercury (mmHg)
  • Centimeters of water (cmH2O)
  • Per minute (#/min)
  • Milliliters per minute (mL/min)
  • Milligram per kilogram (mg/kg)
  • Grams per deciliter (g/dL)
  • Milligram per liter (mg/L)
  • Milligram per deciliter (mg/dL)
  • Nanogram per milliliter (ng/mL)
  • Globules per deciliter (g/dL)
  • Units per liter (U/L)
  • Units per milliliter (U/mL)
  • Milliequivalents per liter (mEq/L)
  • Millimoles per liter (mmol/L)
  • Pack years (pack-yrs)
  • “times two” -> “x2” (1 through 10)
  • “two plus” -> “2+” (1 through 10)
Individual vertebrae (spoken form “letter number”)
  • C1, C2, C3, C4, C5, C6, C7
  • T1, T2, T3, T4, T5, T6, T7, T8, T9, T10, T11, T12
  • L1, L2, L3, L4, L5
  • S1, S2, S3, S4, S5
Sequential, adjacent vertebrae (spoken form “letter number letter number”)
  • C1–C2, C2–C3, C3–C4, C4–C5, C5–C6, C6–C7
  • C7–T1
  • T1–T2, T2–T3, T3–T4, T4–T5, T5–T6, T6–T7, T7–T8, T8–T9, T9–T10, T10–T11, T11–T12
  • T12–L1
  • L1–L2, L2–L3, L3–L4, L4–L5
  • L5–S1
  • S1–S2, S2–S3, S3–S4, S4–S5
Staging for cancer, chronic kidney disease, etc. ()“spoken form” -> written form)
  • “stage zero” -> Stage 0
  • “stage one” -> Stage I
  • “stage one A” -> Stage IA
  • “stage one B” -> Stage IB
  • “stage one C” -> Stage IC
  • “stage two” -> Stage II
  • “stage two A” -> Stage IIA
  • “stage two B” -> Stage IIB
  • “stage two C” -> Stage IIC
  • “stage three” -> Stage III
  • “stage three A” -> Stage IIIA
  • “stage three B” -> Stage IIIB
  • “stage three C” -> Stage IIIC
  • “stage four” -> Stage IV
  • “stage four A” -> Stage IVA
  • “stage four B” -> Stage IVB
  • “stage four C” -> Stage IVC
See examples below for more information on how numbers and measurements/units configurations work together

Numeric Ranges

Note: Numeric range formatting supported for integers only (not numbers with decimals).

Ordinals


Examples

The following examples show expected output given different combinations of formatting configurations.
“the injury was sustained on fifth of February twenty twenty six and surgical repair was completed on February tenth two thousand twenty six”
Notice that, while using locale parameters, the date output is normalized for two different spoken form patterns in the example text
“zero two zero three twenty twenty five”
Dictation of numbers to be shown in “short” date form is not supported. In this example the expected output is based on default numbers:numerals_above_nine formatting.For text to be formatted as a date, dictate the month, day, year pattern as shown above.
“the child had one doctor visit today and is one meter tall”
“one eighty centimeters”
“one hundred eighty centimeters”
Notice that both forms of “hundreds” are normalized using numerals_above_nine and numerals
“ninety eight point two percent”
“First time he tried he came in tenth place”
“she vomited two to three times today”
“one point five to two point five”
“The pain is a six out of ten”

While these examples are based on primaryLanguage:en, the same logic applies for other languages with formatting support. Please contact us for further assistance with formatting configuration.