Create a python function to detect if audio recording uses full band
Budget: $250 – $750 USD
I need a python script that can run on audio files (wav, mp3, m4a, ogg, flac) with different sampling frequency. the function needs to output
- a 0/1 value whether the audio content uses full band of the file sampling frequency or is band limited.
- if band limited - provide the cut-off frequency where no more energy is present.
- provide a 128-long vector (linear-scaled) of energy per frequency band (out of input sampling frequency).
The function should work for inputs sampled at 16kHz, 24kHz, 36kHz, 44.1kHz, 48kHz and 96kHz. it should also handle typical audio resolutions (16bit, 24bit, 32bit + float)
The function should down-mix the input to mono before analysing if needed.
the function should only use code that can be used commercially without acquiring any additional licenses.
Input files will be less than 30s long.
the function needs to run on 10 files of 30s in under 1s (assuming audio is loaded from a fast (e.g. NVMe) disk).
the supplier will need to sign an IP release form, so that we own the created code.
- a 0/1 value whether the audio content uses full band of the file sampling frequency or is band limited.
- if band limited - provide the cut-off frequency where no more energy is present.
- provide a 128-long vector (linear-scaled) of energy per frequency band (out of input sampling frequency).
The function should work for inputs sampled at 16kHz, 24kHz, 36kHz, 44.1kHz, 48kHz and 96kHz. it should also handle typical audio resolutions (16bit, 24bit, 32bit + float)
The function should down-mix the input to mono before analysing if needed.
the function should only use code that can be used commercially without acquiring any additional licenses.
Input files will be less than 30s long.
the function needs to run on 10 files of 30s in under 1s (assuming audio is loaded from a fast (e.g. NVMe) disk).
the supplier will need to sign an IP release form, so that we own the created code.