Is it possible to set a 1d array to a tensorflow 2 model with tflite?

I am trying to use a Keras model saved in tflite in C#. For preformance constraints, I would like to have a 1D array as an input to my model.

But Keras always seems to add a dimension before for the batch size:

import tensorflow as tf

input_layer = tf.keras.layers.Input(128)

Model: "model_2"
__________________________________________________________________________________________________
Layer (type)                    Output Shape         Param #     Connected to                     
==================================================================================================
input_1 (InputLayer)            [(None, 128)]        0    

I then export it to tflite using the following:

converter = tf.lite.TFLiteConverter.from_keras_model(keras_model)
model_lite = converter.convert()
with open("model.tflite", "wb") as f:
  f.write(model_lite)

When I upen my tflite model, the dimension is (1, 128) I haven't found a way to make it 128

Here it is None, I can set it to 1 but the first dimension remains, I am not sure how to discard it, there seems to be no way to do this in the tensorflow documentation.

I use: tensorflow-2.6.5 and python 3.7.5