Missing attribute error whenever torchaudio.save is run

[oganesso]

Whenever I run torchaudio.save('generated.wav', gen.squeeze(0).cpu, 24000) I always gets an error of the "src" object in the "torchaudio" module not having any attributes!

AttributeError                            Traceback (most recent call last)
~\AppData\Local\Temp\ipykernel_15111\511151117.py in <module>
      7 gen = tts.tts_with_preset(text, voice_samples=voice_samples, conditioning_latents=conditioning_latents, 
      8                           preset=preset)
----> 9 torchaudio.save('generated.wav', gen.squeeze(0).cpu, 24000)
     10 IPython.display.Audio('generated.wav')

c:\ProgramData\Anaconda3\lib\site-packages\torchaudio\backend\soundfile_backend.py in save(filepath, src, sample_rate, channels_first, compression, format, encoding, bits_per_sample)
    421             "Using 16 bits per sample might be able to avoid this."
    422         )
--> 423     subtype = _get_subtype(src.dtype, ext, encoding, bits_per_sample)
    424 
    425     # sph is a extension used in TED-LIUM but soundfile does not recognize it as NIST format,

AttributeError: 'builtin_function_or_method' object has no attribute 'dtype'

It doesn't matter what it is, whether that be "dtype" or "ndim"!...

Even if I hypothetically inserted a line that included "src.[anything]" before line 423, I would get an error of no attribute😭