Glossary
The International Phonetic Alphabet (IPA) is the standardized notation system for representing the sounds of any human language. Created in 1888 and refined ever since, IPA gives every distinct sound (phoneme or phone) a unique symbol. Linguists, speech-language pathologists, dictionary editors, and second-language learners all use IPA. ASR engineers use it tangentially — most modern systems don't operate on IPA directly, but it's the universal reference for talking about speech sounds.
How IPA works
Each IPA symbol represents a single sound, regardless of spelling. English "shoe" /ʃuː/, French "chou" /ʃu/, German "Schuh" /ʃuː/ — all start with the same phoneme /ʃ/, even though spelling varies wildly. IPA covers ~107 letters, ~52 diacritics, and ~31 suprasegmentals (stress, tone, length). Together they can transcribe any sound in any language unambiguously.
IPA in dictionaries and ASR
Most English dictionaries (OED, Merriam-Webster, Cambridge) include IPA pronunciations next to each headword. Pronouncing dictionaries like CMU Pronouncing Dictionary use ARPABET (an ASCII-friendly variant) instead of IPA Unicode for engineering convenience. ASR research papers cite IPA when discussing acoustic modeling or specific languages. Production ASR systems mostly skip IPA entirely — they go from audio to subword text tokens directly.
Why IPA matters less for modern ASR
Pre-2018 ASR pipelines explicitly mapped audio to phoneme sequences (often in IPA or ARPABET notation) and then composed phonemes into words via a pronunciation lexicon. Modern Whisper-style systems are end-to-end audio→text, with no explicit phoneme intermediate. The model learns its own internal representation that may or may not correspond to IPA categories. IPA remains useful for human analysis but is no longer load-bearing for ASR engineering.
In practice
When linguists annotate a corpus in ELAN with phonetic detail, they use IPA. Example transcription: English "butter" can be /ˈbʌtər/ (American), /ˈbʌtə/ (British RP), or /ˈbʌʔə/ (Cockney glottal stop). Same word, three IPA representations capturing dialectal variation. Useful for dialect studies; irrelevant to standard ASR which just outputs "butter".
Related terms
Further reading
Try AI transcription
Get started — freeLast updated: September 21, 2026
Frequently Asked Questions
A standardized notation system for representing the sounds of any spoken language. Used in linguistics and pronunciation guides.
IPA (International Phonetic Alphabet) affects how you understand transcript quality, timing, compatibility, or the technology behind speech-to-text results.
Use the related workflow at /ai-transcription when you want to see how this glossary concept connects to an actual transcription task.
Browse the full transcription glossary or read the complete guide to AI transcription.