Transcribing German written text to computer readable phonetic alphabet

Question

I already found some programs that can transcribe text automatically but they don't comply with my requirements. I need:

• A software that transcribes written text to IPA, SAMPA or some other phonetic representation (command line usable, so that I can use the output)
• A software that is free or, even better, open source
• Not a website but an actual downloadable software
• It should be able to split the tanscription into syllables
• It should be able to deal with words that are not in a dictionary, by approximating a good transcription of the unknown word
• No limitation to the number of words that are transcribed
• German language

Nice to have:

• POS-Tagging

A combination of tools to achieve the requirements would also be appreciated.

Answer 1

espeak does a lot of what you are after, it seems to deal with both arbitrary words as well as having a list of exceptions. Though how complete this list of exceptions is remains to be seen.

espeak -q -v de --ipa warm

It seems to be the case that german orthography is pretty regular... though I don't fluently speak german so don't trust me.

Worth noting: IPA is a useful approximation, a more complete model is a 2-d space plus roundedness. To get an idea how much of a lie IPA can be have a look at this chart of korean vowels.

Also worth noting there is controversy over whether the /ɛː/ sound even exists as in hähnchen.

Depending on your use cases you might prefer to use google text-to-speech (commmand-line tool here ). This, unfortunately, provides no ipa output but it is pretty good, so if you just want to know how something is pronounced and have learned to hear all the german vowels then this can be a good reference tool.