luc-vocab
commented
2 months ago You can already do this if you select the portion of the word which exists on forvo, and generate the audio for that selection, do you know how to do it ?
Open mtthth opened 10 months ago
luc-vocab
commented
2 months ago You can already do this if you select the portion of the word which exists on forvo, and generate the audio for that selection, do you know how to do it ?
outercript
commented
2 months ago @luc-vocab I understood the request the other way around.
1) For a given "word", get an audio for a "sentence" containing such word. e.g., input word "house" -> audio for "my house is nice"
2) For a given "sentence" get an audio for either: a) the exact sentence "i am really happy" -> "i am really happy"
b) another sentence that contains the input sentence. "i am really happy" -> "this morning, i am really happy"
luc-vocab
commented
2 months ago ah understood, thought this would require UI improvements, because you'd need to select the sentence manually, it's probably out of the question to add all sentences automatically.
It would be very useful if we were able to download phrase which contains a particular word from Forvo.
For example, looking for "hitherto" would return the prononciation of "A new conscience for truths that have hitherto remained unheard."
It would be also useful to obtain by this way an audio sample which is not directly accessible. For example "I'm really happy" is not in the Forvo base, but the phrase "We've made some progress on this project today, so I'm really happy. Let's celebrate!" is available.