Affected: tensorflow

TensorFlow is an open source platform for machine learning. Prior to versions 2.12.0 and 2.11.1, there is a floating point exception in AudioSpectrogram. A fix is included in TensorFlow version 2.12.0 and version 2.11.1.

Affected: PyPI: tensorflow, PyPI: tensorflow-cpu, PyPI: tensorflow-gpu

### Impact version:2.11.0 //core/ops/audio_ops.cc:70

Status SpectrogramShapeFn(InferenceContext* c) { ShapeHandle input; TF_RETURN_IF_ERROR(c->WithRank(c->input(0), 2, &input)); int32_t window_size; TF_RETURN_IF_ERROR(c->GetAttr("window_size", &window_size)); int32_t stride; TF_RETURN_IF_ERROR(c->GetAttr("stride", &stride)); .....[1]

DimensionHandle input_length = c->Dim(input, 0); DimensionHandle input_channels = c->Dim(input, 1);

DimensionHandle output_length; if (!c->ValueKnown(input_length)) { output_length = c->UnknownDim(); } else { const int64_t input_length_value = c->Value(input_length); const int64_t length_minus_window = (input_length_value - window_size); int64_t output_length_value; if (length_minus_window < 0) { output_length_value = 0; } else { output_length_value = 1 + (length_minus_window / stride); .....[2] } output_length = c->MakeDim(output_length_value); }

Get the value of stride at [1], and the used at [2] ```python import tensorflow as tf

para = {'input': tf.constant([[14.], [24.]], dtype=tf.float32), 'window_size': 1, 'stride': 0, 'magnitude_squared': False} func = tf.raw_ops.AudioSpectrogram

@tf.function(jit_compile=True) def fuzz_jit(): y = func(**para) return y

fuzz_jit() ```

### Patches We have patched the issue in GitHub commit [d0d4e779da0d0f56499c6fa5ba09f0a576cc6b14](https://github.com/tensorflow/tensorflow/commit/d0d4e779da0d0f56499c6fa5ba09f0a576cc6b14).

The fix will be included in TensorFlow 2.12.0. We will also cherrypick th…