Dictionaries that talk: a new OED feature
This month, the Oxford English Dictionary has started to talk. With the addition of audio pronunciations, OED joins other titles in OUP’s family of dictionaries. OED users can now look up a word and click the new blue button next to each pronunciation transcription to hear the world spoken aloud by a British and an American speaker. There’s more detail in the more recent release notes.
For many people, listening to a spoken pronunciation will be a more direct and accessible experience than interpreting a transcription. Does adding audio mean that transcriptions are no longer needed? No, as this extract from my contribution to the new Oxford Handbook of Lexicography explains:
“A user who is in a quiet place, or who has hearing difficulties, or who for whatever reason requires the pronunciation in written form, will continue to look for a transcription. If synthesis is used to produce audio files, transcription would be required as its input; if actors are used to record words aloud, they too require accurate transcriptions to read. A recording made by a speaker will also, of course, reflect that individual’s particular speech, whereas a phonemic transcription is a broader abstraction which, properly interpreted, allows multiple potential variations in accent and idiolect to be embraced.”
Nevertheless, adding audio is an exciting enhancement and a step forward for OED.
How is the audio prepared, and who are the speakers? Not all of the recordings were created afresh for the OED, since of course many of the words in OED also appear in other OUP dictionaries which already have audio, much of which has been recorded under the same conditions and to the same standards. But after a complicated matching-up exercise, we were still left with plenty of work to do. We recruited people who were happy behind a microphone and who combined an ability to read our pronunciation transcriptions fluently with a clear voice, a suitable accent, and meticulous attention to detail. Gradually we built up a collection of sound files. The phonetic skills of our recording artists were vital, as they and our sound engineer checked and made fixes to tiny errors as they went along. This video clip offers a glimpse into the process.
At this point, our American speaker Clifford has stepped out of the sound booth, where he has been reading a list of transcriptions. He is now listening back with our sound engineer Gary, who is segmenting the recording into single named files as they go along. You can hear them select the latter of two versions of the word accumulativeness.
The process of recording and adding more audio will continue into 2016.