2

There should be too many native speakers here, How do you rate TTS (Text to speech) engines on mobile and desktop machines and also there are TTS online like google translate. If someone use them to learn pronunciation of words, could it lead to bad habits and they should be avoided? If yes then what is your recommendation?

closed as primarily opinion-based by IQV, Björn Friedrich, Philipp, Hubert Schölnast, Volker Landgraf Nov 1 '18 at 10:17

Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 4
    As a learner, I strongly recommend against them. Why using them while you have a better option?! There are several dictionaries that provide you with a natural human voice instead (e.g., Duden, PONS, dict.cc,...etc.). There are many aspects of pronunciation that are not caught by such engines such as words that have multiple pronunciations (e.g., seperable-inseperable verbs), assimilation, intonation, among others. The best option IMO is to learn the phonemes and the IPA to be able to read the pronunciation. The previous skill is invaluable in English and German for advanced learner. – Abdullah Oct 31 '18 at 9:56
  • @User among all those alternatives software with natural human voice which one do you recommend for offline use. I plan to learn IPA as well. – pouya Oct 31 '18 at 10:39
  • 2
    Generally, Duden is considered the best. Regarding pronunciation, I think they are almost identical, so try to get anyone offline and it will be fine. Way to go! I recommend to learn the IPA as early as possible :) – Abdullah Oct 31 '18 at 11:02
  • 1
    The only tricky part of pronouncing standard German is identifying reduction syllables and TTS systems are notoriously bad at this. Especially the existence of the Tiefschwa vowel is often denied. It may swallow whole groups of consonants around one or more vocalic r. TTS systems often ignore this and produce an output which is easily identified as "computer voice". – Janka Oct 31 '18 at 13:52
  • @Janka ... which we also call a Roboterstimme (including the Tiefschwa) – Christian Geiselmann Oct 31 '18 at 17:21

Browse other questions tagged or ask your own question.