⌨️SAMPA — the Speech Assessment Methods Phonetic Alphabet

SAMPA is a machine-readable phonetic alphabet: it maps IPA symbols onto printable 7-bit ASCII (codes 33–127), created so phonetic transcription could survive the computing of the late 1980s — e-mail, databases, and speech-technology systems that could not render IPA glyphs.¹ Crucially, SAMPA is not one universal mapping but a family: a core mapping plus per-language transcription guidelines, developed language by language with native speakers.¹

Origins — with an honest asterisk

Per the official UCL page, SAMPA was "originally developed under the ESPRIT project 1541, SAM (Speech Assessment Methods) in 1987–89 by an international group of phoneticians."¹ Wells's own 1995 X-SAMPA paper instead says 1988–91 under ESPRIT project 2589 — a genuine discrepancy in the primary materials, reported here rather than silently resolved.² Either way, the collaborative character is the point: unlike X-SAMPA or Kirshenbaum (single-author schemes), SAMPA "represents the outcome of collaboration and consultation among speech researchers in many different countries."¹

Language coverage grew in waves: the initial six by 1989 (Danish, Dutch, English, French, German, Italian); Norwegian and Swedish by 1992; Greek, Portuguese, Spanish in 1993; Bulgarian, Estonian, Hungarian, Polish, Romanian under BABEL (1996); Arabic, Hebrew, Turkish under OrienTel; plus Cantonese, Croatian, Czech, Russian, Slovenian, Thai.¹

The design in one rule — and a taste of the table

Base rule: IPA symbols that coincide with lower-case Latin letters stay themselves; everything else is recoded within ASCII 37–126. The scheme is case-sensitive and uniquely parsable with no spaces between symbols.¹

Two official obsolescence notes bite anyone reading old materials: the original SAMPA tone marks (` falling, ' rising) were based on pre-1990 IPA and are superseded by the companion prosody system SAMPROSA; and SAMPA originally wrote the syllabicity diacritic before its base (=n), whereas ISO/Unicode practice now requires diacritics to follow the base.¹

Where SAMPA stands today

SAMPA remains embedded in legacy speech-technology resources (EUROM 1, BABEL, Onomastica, OrienTel; used by Oxford University Press; listed by the LDC).¹ For new work, Unicode has removed the original motivation and X-SAMPA covers universal needs — but SAMPA literacy still matters whenever you open an older pronunciation lexicon or speech corpus.

Errors to avoid

• Treating SAMPA as one universal alphabet — the same ASCII character can map differently across the per-language tables.¹
• Conflating SAMPA (per-language) with X-SAMPA (one universal mapping).²
• Using the obsolete tone marks or the pre-Unicode diacritic order.¹
• Expecting prosody coverage — that is SAMPROSA's job.¹

Learn more

• Official SAMPA home page (UCL) — The canonical reference — read in full for this guide.
• Wells (1997), "SAMPA computer readable phonetic alphabet" (Handbook chapter) — The canonical printed citation (Gibbon, Moore & Winski, eds., Mouton de Gruyter).
• X-SAMPA — our companion guide — The universal extension of SAMPA.