Glossary
Spoken language identification
Spoken language identification is the task of working out which language is being spoken in a recording. Products call it automatic language detection. It usually chooses from a short list of candidate languages and decides for a segment of audio rather than word by word.
Updated October 17, 20264 cited facts, each traced to its source
Also called language identification, LID, automatic language detection.
Spoken language identification, often shortened to LID, is the step that works out which language is being spoken in a recording. Products call the same step automatic language detection. A recognizer can run LID first and then load the matching model, or it can check the language as it transcribes. Either way, the result decides which vocabulary and sound patterns are used to write the words down.
Detection works from a list of candidate languages
Most systems do not consider every language on earth. They compare the audio against a set of candidate languages, and the list is limited. Azure Speech's continuous language identification accepts up to 10 candidate languages, while its at-start mode accepts up to four 1. A narrow list makes the choice easier, and a wrong list can force a wrong answer.
A detector can only pick languages it was trained on
A detector cannot recognize a language that never appeared in its training data. The Whisper paper says its training data contains no speech for 20 of the 102 languages in the FLEURS benchmark, and the paper says this caps its language identification accuracy at 80.4% 2. For speakers of languages with little training audio, detection is therefore less reliable, and so is the transcript that follows it.
Detection usually decides per segment, not per word
Most systems decide once for a stretch of speech. The faster-whisper library, for example, has a setting that controls how many segments are considered for language detection 3. A detector that decides once for a segment will not notice a switch inside that segment. Azure's continuous mode behaves the same way, as the code-switching entry explains.
Why it matters for dictation
If detection chooses the wrong language, the transcript is often a set of near-miss words in the wrong language rather than a clean error. Google says automatic language detection is turned on by default in Gboard's advanced voice typing, on supported languages 4. That default helps when you switch between sessions. It is less useful inside one sentence. If you always dictate in one language, set it by hand, because a fixed choice removes one source of error.
How it relates to other terms
Detection is a separate step from transcription. A tool can identify the language correctly and still transcribe it badly, because the accuracy of the recognizer is a different measurement. For that, see word error rate, and for languages with little data, see low-resource language.
Related terms
- Code-switching: speech that moves between languages.
- Low-resource language: a language with little training data.
- Automatic speech recognition: the recognizer that follows detection.
Guides 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.
- 2.
- 3.
- 4.
Related terms
- Dictation Languages, Accents and Voice TranslationThe topic this term belongs to
- AccentAn 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.
- 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.