How to transcribe an aria into IPA: a step-by-step method for singers

An aria in a language you do not speak needs a pronunciation plan before it needs a tempo. The IPA is how you write that plan down. This page sets out the choices you make when you transcribe, and the ready-made sources that save you the work. The full set of language guides is in the diction section.

Why singers transcribe

Classically trained vocalists use the IPA to assist with learning diction in languages other than their own (Ball State University Libraries guide)¹. A transcription of the sung text turns spelling into sounds you can name. For most languages, phonetic transcription shows pronunciation with something much nearer to a one-to-one relationship between sound and symbol than the spelling allows². It lets you step outside the spelling and examine differences in pronunciation between dialects of a language². If you are new to the symbols, start with IPA for singers.

The International Phonetic Association promotes the scientific study of phonetics, including the development of the IPA¹. The method below turns the principle into a score you can sing from.

How to write a phonetic transcription

Phonetic transcription, or phonetic notation, is the written representation of speech sounds². The most common type uses a phonetic alphabet, such as the International Phonetic Alphabet². Most phonetic transcription is based on the assumption that linguistic sounds are segmentable into discrete units that symbols can represent². It indicates the articulatory and acoustic details of those sounds².

A basic principle is that a phonetic transcription should apply to all languages². A transcription devised for one individual language or group of languages is an orthography, not a phonetic transcription². The steps below cover the choices that remain: how detailed to be, which brackets to use and where to start.

Step 1: choose the kind of transcription

Phonemic transcription distinguishes only the phonemes of the language, disregarding allophonic differences and differences between speakers². Phonetic transcription lies on a continuum between broad and narrow². Broad transcription indicates only the most salient phonetic features, and narrow transcription indicates additional phonetic details to varying degrees². A transcription that gives only a basic idea of the sounds of a language in the broadest terms is a broad transcription². A close transcription, indicating precise details of the sounds, is a narrow one².

Two in-between types exist. A transcription that includes some allophonic detail but still closely follows the phonemic structure of an utterance is an allophonic transcription². A transcription that accommodates the variation between the phonemic systems of different varieties of a language is a diaphonemic transcription². A broad transcription may be considered one of several possible solutions for a phonemic representation².

For a working score, choose broad. A narrow transcription is rarely representative of all dialects or speakers of a language, while broad transcription may cover multiple dialects². Broad transcription is the better fit for the pronunciation data in ordinary dictionaries². Narrow transcription may help learners produce proper pronunciation².

Step 2: choose brackets

Square brackets are placed around phonetic representations, and slashes around phonemic representations². The International Phonetic Association recommends that a phonetic transcription be enclosed in square brackets². A transcription that specifically denotes only phonemic contrasts may be enclosed in slashes². If you are unsure, use brackets: setting a transcription off with slashes makes a theoretical claim². Every IPA table on this site uses brackets.

Step 3: start from a source, not a blank page

Several resources do the first draft for you.

• IPA Source. A fee-based service with literal translations and IPA transcriptions of songs and arias in Latin, Italian, German, French, Spanish and English³. You can view a preview sample for free, or pay to download an individual song or aria³. With it, educators and artists can skip the translating and phonetic work and sing with confidence and accuracy¹. Its catalog has amassed 15,804 professionally translated and transcribed titles¹.
• SingersBabel. It lets you hear a native speaker reciting the text while you see the original text, a word-for-word English translation and an IPA guide¹.
• Published collections. Emory’s guide to translations lists online resources with word-for-word translations of art songs and opera arias, plus IPA transcriptions, some with spoken-word audio files of selected texts⁴. An anthology there gives IPA transcriptions uniquely designed to match the musical setting⁴. A book there includes multiple practical exercises in IPA transcription, covering vowels, diphthongs and fricatives⁴. Another book gives a thorough discussion of the method of transcription in its appendix⁴. Some song resources include enunciation and transcription charts for each language on a single page⁴. One title includes a glossary of phonetic and anatomical terms, tongue twister exercises and a suggested repertoire for lyric diction study⁴.
• Anthologies and series. The Diction Coach series includes recitation of texts, IPA and translations⁵. Leyerle song anthologies usually have texts, IPA and translations⁵.
• Pronunciation dictionaries. Pronunciation dictionaries give transcriptions for over 10,000 names of composers, poets, artists, roles, performers, characters and places¹. The Longman Pronunciation Dictionary, third edition, gives over 225,000 pronunciations in both British and American English, in IPA⁵. One dictionary of musical terms gives phonetic readings but no IPA, so do not expect symbols from it⁵. A keyword search in the IPA collection finds transliterations of songs, arias or complete operas¹.
• Software and fonts. PhoneticSoft is paid software to create IPA on your PC for Latin, Italian, German and French texts⁵. A free download of an IPA palette produces documents with IPA fonts⁵.
• Books by language. A book of phonetic readings covers 413 Italian, German and French lyrics¹. A Czech diction book uses the IPA to take the reader systematically through each sound of the language, and covers approaches to singing double consonants, stress and length¹. It includes texts to major song literature and opera excerpts by Smetana, Dvořák, Janáček, Martinů and Haas¹. A choral primer brings together 30 essays by experts from around the world on how the symbols of the IPA can be used¹. Its musical examples illustrate the rhythmic timing and release of consonants within the transcriptions¹. Phrasal and emphatic stress are among the often-overlooked topics explored in one reference¹.
• Library shelves. Browsing call number MT883 in a library turns up language-specific guides that give line-by-line IPA transcriptions, word-for-word translations and phonetic breakdowns for classical vocal literature³.
• The chart. The IPA chart of the International Phonetic Association is the reference for the symbols³, and the Association reissues it annually¹.

Step 4: take care with automatic converters

A converter that provides IPA for Chinese, English, French, Italian, Japanese, Russian and Spanish gives it as spoken, not necessarily as sung¹. An anthology on Emory’s list offers IPA uniquely designed to match the musical setting, which is a different job from converting spoken text⁴. Check converter output against a source made for singing, such as the libraries above^1,3.

Step 5: apply the language’s rules

A transcription is only as good as the rules behind it. Italian gives a compact example from a Stanford primer for singers.

• Italian has only 7 vowel sounds⁶.
• Italian vowels are pure⁶. Unaccented English vowels tend to change toward a more neutral sound, and no Italian vowel ever makes the schwa sound⁶.
• Open e makes a sound similar to the English short e, and open o is like the vowel in awe⁶.
• Italian unstressed e and o are always closed⁶.
• Diphthongs are always written with two vowels and are frequent in Italian⁶.
• Double consonants take noticeably longer to say than single ones, and the vowel before them takes less time⁶.
• The r is flipped between two vowels⁶.

No Italian vowel ever makes the schwa sound, so a schwa in an Italian line is a mistake⁶. A doubled consonant in the spelling should show as a longer sound⁶. The full rule set is on the Italian diction page; the other languages are covered on their own pages in the diction section.

Step 6: check with an ear

SingersBabel pairs a native speaker’s recitation with the original text, a word-for-word English translation and an IPA guide¹. Listen repeatedly and attentively to good spoken and sung Italian⁶. What a diction coach does explains what a coach adds to that work. For a worked language, see Italian diction for singers, where the rules behind each symbol are laid out.

Coursework and library guides

A university diction course teaches the same skill. The University of Arkansas - Fort Smith syllabus for Diction for Singers I lists reading IPA transcriptions of English and Italian words as a course outcome⁷. Another outcome is using the IPA as a tool to achieve clarity and accuracy of pronunciation in singing⁷. At the Musicians Institute library guide, IPA Source is listed as not free, with thousands of IPA transcriptions for primarily classical arias, as well as translations⁸.

A short checklist

• Choose broad transcription and square brackets².
• Begin from IPA Source, a published collection or a recording with a native speaker^1,3,4.
• Check converter output against a source made for singing¹.
• Run the language’s vowel and consonant rules over the result⁶.