Thursday 7 October 2010

English Transcription Editor v1.7

The English Transcription Editor PhoTransEdit is a free Windows tool to make written transcriptions easier. All you have to do is to enter the text to be transcribed in regular orthography and to press the button 'Transcribe'. If all the words are found in the database provided, the transcribed text appears on the screen. You can then copy and paste it into your word processor.
Here's a screenshot: is where you can read more about it and also download the installation file.


  1. Jack-of-all-trades7 October 2010 at 19:43

    The mistake is, I think, that the weak form of "to" is /tu/ (for me, [tʊ]) before a vowel, but /tə/ before a consonant (except "in very formal speech": J. Ch. Wells, "Longman pronunciation dictionary", under the entry "to"). It should be /tə bi/.

    By the way, I hate -- well, I strongly dislike, if you prefer -- these technical devices. They usually make a lot of mistakes, especially in the field of syntactic phonetics: they tend to ignore assimilations, stress shifts, etc. And they easily confound homographs, of course.

    As for me, I prefer to cultivate the vice of thinking with my own head. Writing phonetic transcriptions with my pen is a good exercise, after all.

  2. Jack-of-all-trades7 October 2010 at 20:12

    If you write "To be or not to be? That is the question", here's what you get:

    | tu bi ɔː nɒt tu bi | ðət s ðə ˈkwestʃən |

    It should be /tə bi/, again, and /ðæt ɪz/.

    The device is of course unable to distinguish the various senses of "that", and can't know that this particular "that", being a demonstrative, has no weak form.

    It is also unable to distinguish different contexts and styles, and doesn't know that here the strong form /ɪz/ sounds much better than the weak form /s/.

    It is not to say that such tools are completely useless; but they should be used, I think, with caution (and sparingly).

  3. @Jack-of-all-trades:
    "It is not to say that such tools are completely useless; but they should be used, I think, with caution (and sparingly)." How true!

  4. The stress marking seems a bit hit and miss too. /ðɪs/, /tekst/ and /spɒt/ seem likely candidates for rhythmic stress in your example.

  5. Jack-of-all-trades9 October 2010 at 10:37

    Yes. And some strange transcription mistakes sometimes: for "Jerome's", the device gives

    | dʒeˈrəʊmɪz |

    If you write "lover", you get | ˈlʌvə |; but if you want to know the pronunciation of the Saxon genitive, for "lover's" you get

    | ˈləvərz |

    (with an /r/ that's impossible here, in the received pronunciation, and with /ə/ instead of /ʌ/)!

  6. All well observed, dear blog readers!
    Just to make sure, let me add two remarks:
    (1) My blog entry was and is not intended as an advert for the editor.
    (2) I previously used (and still use) the 'Unicode Phonetic Keyboard', which you can find on the webpages of University College London. It's a nice tools if you have to insert a few IPA signs only into an otherwise normal text. But this method is fairly clumsy when it comes to transcribing larger stretches. So I thought it might be less time-consuming if one uses the transcription editor and edit the result.

  7. Excellent and informative blog. Thanks for posting this. It’s informative and enlightening. Keep up the good work.

    Free Transcription

  8. @mahesh: Thanks a lot for your compliment!