Whisper compute_type int8 value is not taken

[thepenguinthatwants]

Using docker-compose and compose-gpu.yml

edited Settings.py line to be

whisper_compute_type = int8 # change to "int8" if low on GPU mem (may reduce accuracy)

But gotten this error

ValueError: Requested float16 compute type, but the target device or backend do not support efficient float16 computation.
root@7a558ceaeb86:/app/UltraSinger/src# python UltraSinger.py --interactive
2025-02-15 17:43:30.548531: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2025-02-15 17:43:31.354848: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Traceback (most recent call last):
  File "/app/UltraSinger/src/UltraSinger.py", line 12, in <module>
    from modules.init_interactive_mode import init_settings_interactive
  File "/app/UltraSinger/src/modules/init_interactive_mode.py", line 3, in <module>
    from Settings import Settings
  File "/app/UltraSinger/src/Settings.py", line 12, in <module>
    class Settings:
  File "/app/UltraSinger/src/Settings.py", line 46, in Settings
    whisper_compute_type = int8   # change to "int8" if low on GPU mem (may reduce accuracy)
NameError: name 'int8' is not defined. Did you mean: 'int'?

Any clue how to make it work?