Audible instructions Discussion View history

Revision as of 19:02, 19 November 2018 by Kartografer (talk | contribs) (redesigned lead to distinguish tts and non-tts voices, and added non-tts section)

When navigating in Waze with all sound turned on (rather than alerts only), Waze will issue various audible instructions at various times in order to prepare and inform the user to make a certain movement.

The occurrence, timing and format of these instructions do not depend upon the language of the voice, but they do depend upon whether the voice chosen to speak the instructions includes street names. Voices that include street names are speech synthesizers, which convert text to speech (TTS). Voices that do not include street names are simple collections of recorded phrases. We'll call these two categories of voices "TTS" and "non-TTS" respectively.

TTS

TTS voices are sophisticated speech engines, which not only say street names, but also place names and more detailed information in general regarding navigation and alerts. These include Jane, which speaks US English; Kate and Thomas, which speak UK English; and also Javier, Marusya and dozens of others in various languages.

From a TTS voice there are four possible audible instructions before a movement. Depending upon speed, distances, and timing, anywhere from one to four of these instructions may be heard.

Occurrence

We will call the different instructions A, B, C, and D based on the order they would appear. However, some instructions may not occur based on a number of conditions. The following is the order of the most common occurrence:

  • D
  • A, D
  • A, C, D
  • A, B, C, D

Instruction A

Instruction A is given a few seconds after the previous movement (or, if the previous movement is onto a ramp, a few seconds after exiting the ramp(s) onto which the previous instruction brought you), unless you are so close to the next movement that there is not enough time to give both instruction A and D (in which case only instruction D is given).

Format depends on distance to next movement.

  • If distance to next movement > 4.35 mi (7.0 km), "Continue for (x hours and) y minutes to [street name]"
  • If distance to next movement < 4.35 mi (7.0 km), "In [distance], [instruction] [on/to] [street name]"
Distance to next movement Example
> 4.35 mi (7.0 km) Continue straight for 15 minutes to Rue Royale
Continue straight for 2 hours and 15 minutes to Exit 123: I-45 N / Bakerstown
< 4.35 mi (7.0 km) In 700 feet, turn left on Brightview Blvd
In 2.1 miles, exit to Exit 98: SR-76 / Chilltown

Instruction B

Instruction B is given at the point where the movement is expected to be 5 minutes away, but only if it's been 15-20 minutes since instruction A.

Format: "In [x] miles, [instruction] [on/to] [street name]"

examples
In 3.4 miles, turn left on Brightview Blvd
In 6 miles, exit right to Exit 98: SR-76 / Chilltown

Note: while the timing of the instruction is time-based, the instruction is given in distance.

Instruction C

Instruction C is given at a certain distance from the movement depending on your speed as detected at that point, unless the prior instruction was given too close to the timing of instruction C. The street name may or may not be included, depending on how long it has been since Instruction A (a couple minutes).

Format: "In [distance], [instruction] ([on/to] [street name])"

speed range distance example
0 – 43.5 mph (0 – 70 km/h) 1000 ft In a thousand feet, turn right
43.5 – 55.9 mph (70 – 90 km/h) 0.5 mi In half a mile, exit left
55.9 – 93.2 mph (90 – 150 km/h) 1 mi In one mile, stay to the right to I-75 N
93.2+ mph (150+ km/h) 2 mi In two miles, exit right

Instruction D

Instruction D is given every time, just before the movement (unless you make the turn before Waze has a chance to call it out).

Format: "[instruction] [on/to] [street name]"

examples
Turn right on Main St
Exit left to Exit 234B: Poydras St / Superdome


Notes

  • Instruction D is (virtually) always given.
  • Instruction A is given only if there will be instruction D and there is enough time for both.
  • Instruction C is given only if there was instruction A and there has been enough time since instruction A.
  • Instruction B is given only if there will be instruction C (and therefore there was instruction A) and there has been enough time since instruction A.

Combined instructions

If, based on the speed of the user, Waze expects a second movement to occur within about 30 seconds after the upcoming movement, it will tack on a second instruction to the end of instructions C and D separated by "then".

examples
In half a mile, exit right, then stay to the left to Exit 26: I-71 N / Cleveland
Turn left on N High St, then turn right on Arcadia Ave

Preposition choice

When a voice is selected that says street names, such as Jane, those names are preceded by "on" or "to" according to a couple factors:

  • Following an instruction to exit or stay to the left or right, Waze always says "to" and the street name.
  • Following an instruction to turn left or right, continue straight, make a U-turn or do anything at a roundabout, Waze says either "to" or "on" dependent upon the road type of the next segment past the instruction.
    • If the next segment is a  Ramp , Waze says "to" and the street name.
    • If the next segment is any other road type, Waze says "on" and the street name.

The word "to" at the beginning of any street name, which is common on ramp names, is never a factor in whether Waze says "to" or "on" and is actually ignored in audible instructions. This means that if the name of a ramp is inherited through an unnamed non-ramp segment, Waze will say "on" and will omit the "to" completely. For example, turning left through a roundabout made of non-ramp segments and onto a ramp with the name "to US-33 W / Lancaster" will produce the audible instruction "at the roundabout, turn left on US-33 W / Lancaster."

Flowchart

Non-TTS

Non-TTS voices are sets of recorded phrases. These include Amy and Nathan, any novelty voice like Boy Band, most other voices in other languages, and any user-recorded voices. The voice recorder contains the extent of what all non-TTS voices can say, and it is viewable in by tapping "record new voice" in the Waze voice menu.

From a non-TTS voice there are three possible audible instructions before a movement. Depending upon speed, distances, and timing, anywhere from one to three of these instructions may be heard.

First instruction

The first instruction is given when enough time and advance distance is available for it. The actual distance from the previous movement to the next movement is rounded up to give the distance used in the instruction.

Format: "In [distance], [instruction]"

speed and/or distance to next movement example
< 0.1 mi In 0.1 miles, continue straight
0.1 mi - 0.25 mi In a quarter of a mile, turn right
0.25 - 0.5 mi In half a mile, at the roundabout turn left
0-43.5 mph (0-70 km/h) and > 0.25 mi
43.5+ mph (70+ km/h) and > 0.5 mi In 1 mile, make a U-turn

Intermediate instruction

An intermediate instruction is given at a certain distance from the movement depending on your speed, unless the first instruction was given too close to the timing of the intermediate.

Format: "In [distance], [instruction]"

Speed Example
0-43.5 mph (0-70 km/h) In 0.1 miles, keep right
43.5+ mph (70+ km/h) In a quarter of a mile, exit left

Final instruction

The final instruction is given every time, just before the movement (unless you make the turn before Waze has a chance to call it out).

Format: "[instruction]"

Examples
Turn left
At the roundabout take the fifth exit

Combined instructions

If, based on the speed of the user, Waze expects a second movement to occur within about 30 seconds after the upcoming movement, it will tack on a second instruction to the end of instructions C and D separated by "and then".

Examples
Turn left, and then turn right
Exit right, and then keep left