- How do I copy the results line by line to Word?
- Why [r] instead of [ɹ]?
- Can I download the speech audio?
How do I copy the results line by line to Word?
For now “Side by side with English text” is your best option if you are going to copy it somewhere else. For best results you will have to break up the original text by lines though.
Why [r] instead of [ɹ]?
The website provides phonemic transcription. Since there is no rolling [r] sound in English language and therefore no confusion between the two, API recommends using ‘r’ for simplicity. Most dictionaries do just that.
Can I download the speech audio?
Not at the moment. The speech you hear is generated by your browser, so potentially there might be extensions capturing browser’s audio, but you will have to search extension repository of the browser you are using for that.
There is a way, using a Chromebook or smart phone. Download an app called Cloud Audio Recorder from Google Play Store. Open CAR app in another and click the start button on cloud audio recorder, before you go to the tab of “tophonetics.” Now, press the play button on the “tophonetics.” Once you finish listening to the pronunciation of the word or sentence, go back to the tab of the cloud audio recording app and click stop. You can download anything you record in this app. It has an HD sound.
I am writing a book for learning foreign language. can I use IPA for pronouncing german, spanish, italian, portuguese and french ? if yes then should I use British or American Phonetics ?
I was wondering if there is an url-scheme for the iOS app. It doesn’t only make sense for me, if I can switch directly from my used SRS Anki with input to your app, as I do it already with online-dictionaries or the Terminology app.
Thanks for the suggestion. I answered it on the app page http://mu-sonic.com/tophonetics/.
HI, great website! Wondering: what’s the source the British annotation? BEEP? Or is it some transformation of the CMU?
Is there available dataset of words and it’s corresponding phonemes ?
Why are some ‘transcriptions’ in red letters and exactly the same as I typed them, e.g., “Barolo.”
It means the engine couldn’t transcribe the word.
Hi!
In the case of words with multiple transcriptions, these can depend on the word’s role (adjective, noun, etc). For example, ‘suspect’ is ˈsʌˌspɛkt if it’s an adjective or noun, but səˈspɛkt if it’s a verb.
Your website makes this distinction, but I looked into the CMUDict and didn’t find where the multiple transcriptions are associated to roles. Do you mind if I ask where you get this data from?
Thanks in advance!
That’s correct, CMU dictionary doesn’t have that data. CMU transcriptions had to be marked up on our side and then the transcription engine selects the best variant where possible. But you would guess that, right? 🙂
I’m considering using CMUDict on a project of my own (not a competing service btw :)) and I would love to have such markups as well, but since they’re the result of your work I won’t be as shameless as to ask you for them.
Your reply was quite helpful, though! Thank you, and congrats on a well done transcription engine!
What is your standard reference for both American and British spelling? Thank you.
I presume your question is about input spelling. We aim to cover both standard spellings in both dialects. At least this is the case when you select “British” in options. American dictionary needs more work in that regard as it generally covers American spellings only at the moment.