Glossary
Accent
An accent is the way a speaker pronounces a language, shaped by where they grew up, their first language or their community. Accents change pronunciation, not vocabulary, and speech recognition error rates can differ sharply from one accent to another.
Updated October 17, 20264 cited facts, each traced to its source
Also called foreign accent, regional accent.
An accent is the way a speaker pronounces a language: the vowels, consonants, stress and rhythm they use. Two people can use the same words in the same grammar and still sound different. A foreign accent reflects the speaker's first language, as when an English speaker with a first language of Spanish pronounces English vowels differently from a native speaker. A regional accent reflects where the speaker grew up, such as Scottish English or Texan English. Both are accents, and both can affect how well a speech recognizer understands you.
An accent changes how words sound, not which words are used
The accent is in the sound. The vocabulary, grammar and spelling of the language stay the same. That is why an accent is different from a dialect, which can have its own words and grammar, and from a locale, which is a setting you choose on a device.
For a speech recognizer, a change in sound can be enough to change the word it chooses. The acoustic evidence for two words may be close, and the system picks the word that best fits both the sound and the surrounding language. An accent that moves the sounds away from the training data pushes that choice toward the wrong word.
Recognizers make more errors on some accents
In the EdAcc study, the best English speech recognition model, trained on 680 thousand hours of transcribed data, averaged a word error rate of 19.7 percent on conversations between friends with many accents 1. The same model had a word error rate of 2.7 percent on US English clean read speech 2. The two test conditions differ in more than accent, since conversation is harder than read speech, so the gap is not all due to accent. The authors also found that every English model they tested dropped in performance on speakers of Indian, Jamaican and Nigerian English 3.
Why accent bias happens: the training data favors some accents
The cause lies mostly in the training data. A 2021 study of a large English speech model, trained on 44.5K hours of English speech, found an accuracy bias in terms of accentual variety, favoring the accents most prevalent in the training corpus 4. A recognizer does not judge how well you speak. It has learned more examples of some accents than others.
A concrete example: a sentence through two accents
Suppose two people say the same instruction, "send the invoice by Friday", one with a regional accent close to the training data and one with a foreign accent that the training data covers less. The first person may see the sentence typed correctly on the first try. The second may see "invoice" replaced by a similar word, and have to correct it. The sentence is the same, but the recognizer's fit to the sound is not.
Why it matters for dictation and typing
If dictation misses words that you say clearly, do not assume the fault is yours. Choose the locale that matches your region, add names and jargon to a custom vocabulary list, and dictate at a steady pace with the microphone close. Then test the tool on your own speech for a few minutes before you rely on it. Our guide on speech recognition and accents covers the fixes in order, and the question on why dictation does not understand an accent gives the short version.
Related terms
- Dialect: a regional variety with its own words and grammar, which is more than accent.
- Non-native speech: speech in a second language, which usually carries a foreign accent.
- Locale: the language and region setting for a device.
- Word error rate (WER): the measure used to compare recognition errors across accents.
Guides and questions that cover it
Evidence
Every number on this page, with where it comes from. Open a fact to see the exact words in its source.
- 1.On the EdAcc accent test set of conversations between friends, the best-performing English speech recognition model, trained on 680 thousand hours of transcribed data, averaged a word error rate of 19.7 percent, against 2.7 percent on US English clean read speech.
Ramon Sanabria et al., arXiv (accepted to IEEE ICASSP 2023), 2023 · Original
- 2.
- 3.
- 4.
Related terms
- Dictation Languages, Accents and Voice TranslationThe topic this term belongs to
- Character error rate (CER)Character error rate (CER) is the number of character insertions, substitutions and deletions needed to turn a recognizer's output into the correct text, divided by the number of characters in the correct text. It is the usual measure where words are not separated by spaces.
- Code-switchingCode-switching is the habit of moving between two or more languages within a conversation or a single sentence. Speech recognizers built around one language often misread the switched words, so mixed speech is harder to dictate than speech in one language.
- DiacriticsDiacritics are marks added to a letter, such as the acute accent in é, the tilde in ñ or the vowel marks in Arabic, that change how a letter sounds or which word it spells. Speech tools and everyday writers often leave them out, so a dictated word can come out spelled differently from what you meant.
- DialectA dialect is a regional or social variety of a language, with its own vocabulary, grammar and pronunciation. Speech recognition is usually less accurate on dialects than on the standard form, because labeled training data for dialects is often scarcer.
- Input method editor (IME)An input method editor (IME) is software that supports text input for characters that cannot be typed easily on a standard keyboard, such as many East Asian characters. Speech input can also be an input method, and Android allows only one input method to be enabled at a time.
- LocaleA locale is a language combined with a region, and sometimes a script, such as en-GB for English as used in the United Kingdom. Dictation and speech settings use locales because the same language can differ in spelling, punctuation and speech patterns from one region to another.
- Low-resource languageA low-resource language is one with little digital speech and text data for training speech recognition and other language tools. There is no fixed cutoff, and a language can be well resourced for one task and poorly resourced for another.
- Modern Standard Arabic (MSA)Modern Standard Arabic (MSA), also called Fusha, is the standardized form of Arabic used in writing and formal speech, such as news broadcasts and official documents. Speakers use regional dialects in everyday conversation, and speech tools recognize formal Arabic more accurately than dialect speech.
- Non-native speechNon-native speech is speech in a language that is not the speaker's first language, often called L2 speech or foreign-accented speech. Recognizers can transcribe read non-native English well, but they handle hesitations, repetitions and revisions differently, and results depend on the system.
- PinyinPinyin, also called Hanyu Pinyin, is the system that spells Mandarin Chinese syllables with the Latin alphabet and marks the tones with diacritics. People type Chinese by spelling the sound in Pinyin and choosing the character they want, and speech tools must choose among characters that share one Pinyin spelling.
- Right-to-left (RTL) scriptA right-to-left (RTL) script is a writing system, such as Arabic, Hebrew, Persian or Urdu, whose letters run from right to left. Text that mixes these letters with digits or Latin words is bidirectional, so software needs a specific algorithm to order and display it.
- Speech translationSpeech translation converts speech in one language into the same meaning in another language, as text or as synthesized speech. Most systems first recognize the speech and then translate the text, so a recognition error carries into the translation.
Pages that use this term
- Dictation Languages, Accents and Voice TranslationTopic
- Speech Recognition and Accents: Accuracy and FixesGuide
- Spanish Speech to Text and Talk to TextGuide
- Dictation for Language Learning: How and Why It WorksGuide
- Speech Recognition for Stuttering and DysarthriaGuide
- Can dictation help non-native English speakers write work emails?Question
- Can you use voice typing to practice pronunciation?Question
- What is the best dictation app for learning a language?Question
- DiacriticsGlossary
- DialectGlossary
- LocaleGlossary
- Modern Standard Arabic (MSA)Glossary
- Non-native speechGlossary