diff --git a/.github/workflows/ljspeech.yml b/.github/workflows/ljspeech.yml
index b8de39547..19c8f34e1 100644
--- a/.github/workflows/ljspeech.yml
+++ b/.github/workflows/ljspeech.yml
@@ -30,8 +30,8 @@ jobs:
         id: set-matrix
         run: |
           # outputting for debugging purposes
-          python ./.github/scripts/docker/generate_build_matrix.py
-          MATRIX=$(python ./.github/scripts/docker/generate_build_matrix.py)
+          python ./.github/scripts/docker/generate_build_matrix.py --min-torch-version "2.3"
+          MATRIX=$(python ./.github/scripts/docker/generate_build_matrix.py --min-torch-version "2.3")
           echo "::set-output name=matrix::${MATRIX}"
 
   ljspeech:
diff --git a/egs/ljspeech/TTS/README.md b/egs/ljspeech/TTS/README.md
index c9cfc22fd..f5495eeaf 100644
--- a/egs/ljspeech/TTS/README.md
+++ b/egs/ljspeech/TTS/README.md
@@ -176,6 +176,15 @@ The above command generates the following files:
 
 where the 2 in `model-steps-2.onnx` means it uses 2 steps for the ODE solver.
 
+**HINT**: If you get the following error while running `export_onnx.py`:
+
+```
+torch.onnx.errors.UnsupportedOperatorError: Exporting the operator
+'aten::scaled_dot_product_attention' to ONNX opset version 14 is not supported.
+```
+
+please use `torch>=2.2.0`.
+
 
 To export the Hifigan vocoder to onnx, please use: