Are there any other input forms than filenames or microphone?

[NienkeWessel]

I asked this question on stack overflow and they told me to go here. If that is not appropriate, please do tell.

I want to use this library to parse different audio files. However, because of all the preprocessing we need to do, I cannot directly put the audio files into the SpeechRecognition library with sr.AudioFile(). Instead, we have chunks in memory, each consisting of a sentence. I cannot find any way to put in sound except for using the aforementioned AudioFile() or the Microphone() function. It seems unlikely to me that these are the only possibilities, but I cannot seem to find how to do it differently. So my question is: is there any other way to put in the sounds?

The file chunks are currently just in numpy array format, but it should not be difficult to put them into most other sound object formats. So any other possibility to put into the SpeechRecognition library would probably be fine for me.

Steps to reproduce

Not applicable

Expected behaviour

A function to put in a datastructure such as for example a numpy array, but also other sound objects (such as those from librosa) into the library

Actual behaviour

Not applicable

System information

My system is Ubuntu 20.04

My Python version is 3.8.5.

My Pip version is 20.0.2.

My SpeechRecognition library version is 3.8.1.

I don't have PyAudio installed.