The spoken language identifier is a service that tries to determine the language spoken in an audio recording. The model currently supports 8 languages: English, Spanish, Italian, French, German, Portuguese, Dutch, and Russian. You can use it to classify recordings as short as 1 second and as long as a minute. Note that the longer the recording, the higher the accuracy of the prediction. For 20 second recordings the accuracy is about 95%, whereas for 5 second samples it is just over 80%. Supported audio formats: WAV, FLAC, OGG.
This API allows you to extract most relevant terms from a text. It is not, like many others, a basic TF-IDF analysis. It compare the text against a very large language model, it uses a probabilistic model to identify candidates, it supports multi-words terms and not only single words. It uses part of speech tagging to clean up the results". In short it is probably the most advanced term extraction out there.