St. Mary’s is on Church St.
St.Contextual abbreviationAccepted:
- Saint
St.Contextual abbreviationAccepted:
- Street
The Pronunciation Robustness benchmark measures whether Text to Speech models pronounce challenging text correctly, including words whose reading depends on context, shorthand that must be expanded, exact sequences like codes and emails, and standalone terms like brand and place names. Human reviewers listen to each clip and judge highlighted spans as pronounced correctly or not.
Every model reads the same set of sentences. Each sentence carries one or more highlighted spans that are hard to pronounce, with accepted readings defined up front. Human reviewers listen and judge each span.
A fixed sentence set
Sentences across four categories, each with highlighted spans and pre-agreed accepted pronunciations. Every model reads the identical text with a pinned voice.
Every model reads every sentence
Audio is generated once per model with a consistent voice, then served to reviewers in random order without model names. Sentences are sent as written, with no normalization on our side; each model uses its default text normalization setting where it has one.
Human reviewers judge span(s) within sentence
Reviewers hear the clip, see the highlighted span and its accepted readings, and answer whether it was pronounced naturally and correctly with either of Yes, No, or I could not tell. At least three third party reviewers judge over 95% of clips for each published model. A model's score is the share of span judgements answered Yes, out of Yes plus No; "I could not tell" answers are excluded.
Sessions include attention-check clips with known answers. Reviewers who miss any answered check are excluded entirely. We publish results for a model once at least 95% of the samples in the set are covered by three or more reviewers.
Was the highlighted part pronounced naturally and correctly in this sentence?
Press Ctrl+Z to undo the last edit.
Accepted pronunciations: Ctrl+Z
Count it as correct if it matches any option below.
Contextually Appropriate (Contextual Disambiguation)
Words spelled the same but read differently depending on context
e.g. a wound that is bandaged vs. a bandage that is wound
Expanding Shorthand (Text Normalization)
Numbers, dates, units and notation read out naturally
e.g. 6'2", Chapter XVII, or 1 tsp of sugar
Preserving Exact Sequences (Sequence Fidelity)
Codes, paths, emails and identifiers spoken exactly
e.g. .env.local or a.chen@ucsf.edu
Standalone Terms (Term Pronunciation)
Brand, place and technical names pronounced correctly
e.g. Arkansas, façade, or genre
The same sentences, read by every model on the index. Select a model to hear its clips for each category of pronunciation challenge.
Example clips below are a fixed public subset of the benchmark set. The full set is withheld to keep the benchmark uncontaminated.
All example sentences and audio below are for the selected model.
Words spelled the same but read differently depending on context
St. Mary’s is on Church St.
St.Contextual abbreviationAccepted:
St.Contextual abbreviationAccepted:
That excuse does not excuse the delay.
excuseFinal-consonant contrastAccepted:
excuseFinal-consonant contrastAccepted:
The instructions say to wait 30 sec. before reading sec. 4.
30 sec.Contextual abbreviationAccepted:
sec. 4Contextual abbreviationAccepted:
Numbers, dates, units and notation read out naturally
Median latency was 12 ms, reported as p50.
12 msMeasurements and unitsAccepted:
p50Measurements and unitsAccepted:
Take the elevator to the 3rd Fl and turn left.
3rd FlNumbers, ranges and other shorthandAccepted:
The Class of '09 reunion is on 9/9, of all days.
'09Dates and timesAccepted:
9/9,Dates and timesAccepted:
Codes, paths, emails and identifiers spoken exactly
Press Ctrl+Z to undo the last edit.
Ctrl+ZCodes and identifiersAccepted:
We're on iOS 26.1.2 here and macOS 15.7.1 on the laptops.
26.1.2Versions and network valuesAccepted:
15.7.1Versions and network valuesAccepted:
Dial 999 in the UK, 112 in the EU and 911 here.
999Codes and identifiersAccepted:
112Codes and identifiersAccepted:
911Codes and identifiersAccepted:
Brand, place and technical names pronounced correctly
The river continues south through Arkansas.
ArkansasPlace nameAccepted:
The paper was typeset in LaTeX.
LaTeXSoftware termAccepted:
Workers restored the stone façade above the entrance.
façadeBorrowed termAccepted: