diff --git a/whisper-basic/Cargo.toml b/whisper-basic/Cargo.toml
new file mode 100644
index 0000000..06938a2
--- /dev/null
+++ b/whisper-basic/Cargo.toml
@@ -0,0 +1,7 @@
+[package]
+name = "whisper-basic"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+wasmedge-wasi-nn = "0.8.0"
diff --git a/whisper-basic/README.md b/whisper-basic/README.md
new file mode 100644
index 0000000..100d695
--- /dev/null
+++ b/whisper-basic/README.md
@@ -0,0 +1,78 @@
+# Basic Example For WASI-NN with Whisper Backend
+
+This example is for a basic audio recognition with WASI-NN whisper backend in WasmEdge.
+In current status, WasmEdge implement the Whisper backend of WASI-NN in only English. We'll extend more options in the future.
+
+## Dependencies
+
+This crate depends on the `wasmedge-wasi-nn` in the `Cargo.toml`:
+
+```toml
+[dependencies]
+wasmedge-wasi-nn = "0.8.0"
+```
+
+## Build
+
+Compile the application to WebAssembly:
+
+```bash
+cargo build --target=wasm32-wasi --release
+```
+
+The output WASM file will be at [`target/wasm32-wasi/release/whisper-basic.wasm`](whisper-basic.wasm).
+To speed up the processing, we can enable the AOT mode in WasmEdge with:
+
+```bash
+wasmedge compile target/wasm32-wasi/release/whisper-basic.wasm whisper-basic_aot.wasm
+```
+
+## Run
+
+### Test data
+
+The testing audio is located at `./test.wav`.
+
+Users should get the model by the guide from [whisper.cpp repository](https://github.com/ggerganov/whisper.cpp/tree/master/models):
+
+```bash
+curl -sSf https://raw.githubusercontent.com/ggerganov/whisper.cpp/master/models/download-ggml-model.sh | bash -s -- base.en
+```
+
+The model will be stored at `./ggml-base.en.bin`.
+
+### Input Audio
+
+The WASI-NN whisper backend for WasmEdge currently supported 16kHz, 1 channel, and `pcm_s16le` format.
+
+Users can convert their input audio as following `ffmpeg` command:
+
+```bash
+ffmpeg -i test.m4a -acodec pcm_s16le -ac 1 -ar 16000 test.wav
+```
+
+### Execute
+
+> Note: This is prepared for `0.14.2` or later release in the future. Please build from source now.
+
+Users should [install the WasmEdge with WASI-NN plug-in in Whisper backend](https://wasmedge.org/docs/start/install/#wasi-nn-plug-ins).
+
+```bash
+curl -sSf https://raw.githubusercontent.com/WasmEdge/WasmEdge/master/utils/install.sh | bash -s -- --plugins wasi_nn-whisper
+```
+
+Execute the WASM with the `wasmedge` with WASI-NN plug-in:
+
+```bash
+wasmedge --dir .:. whisper-basic_aot.wasm ggml-base.en.bin test.wav
+```
+
+You will get recognized string from the audio file in the output:
+
+```bash
+Read model, size in bytes: 147964211
+Loaded graph into wasi-nn with ID: Graph#0
+Read input tensor, size in bytes: 141408
+Recognized from audio:
+[00:00:00.000 --> 00:00:04.300]  This is a test record for whisper.cpp
+```
diff --git a/whisper-basic/src/main.rs b/whisper-basic/src/main.rs
new file mode 100644
index 0000000..8a5872b
--- /dev/null
+++ b/whisper-basic/src/main.rs
@@ -0,0 +1,35 @@
+use std::env;
+use std::fs;
+use std::error::Error;
+use wasmedge_wasi_nn::{GraphBuilder, GraphEncoding, ExecutionTarget, TensorType};
+
+pub fn main() -> Result<(), Box<dyn Error>> {
+    let args: Vec<String> = env::args().collect();
+    let model_bin_name: &str = &args[1];
+    let wav_name: &str = &args[2];
+
+    let model_bin = fs::read(model_bin_name)?;
+    println!("Read model, size in bytes: {}", model_bin.len());
+
+    let graph = GraphBuilder::new(GraphEncoding::Whisper, ExecutionTarget::CPU).build_from_bytes(&[&model_bin])?;
+    let mut ctx = graph.init_execution_context()?;
+    println!("Loaded graph into wasi-nn with ID: {}", graph);
+
+    // Load the raw pcm tensor.
+    let wav_buf = fs::read(wav_name)?;
+    println!("Read input tensor, size in bytes: {}", wav_buf.len());
+
+    // Set input.
+    ctx.set_input(0, TensorType::F32, &[1, wav_buf.len()], &wav_buf)?;
+
+    // Execute the inference.
+    ctx.compute()?;
+
+    // Retrieve the output.
+    let mut output_buffer = vec![0u8; 2048];
+    _ = ctx.get_output(0, &mut output_buffer)?;
+
+    println!("Recognized from audio: \n{}", String::from_utf8(output_buffer).unwrap());
+    
+    Ok(())
+}
diff --git a/whisper-basic/test.wav b/whisper-basic/test.wav
new file mode 100644
index 0000000..8de174e
Binary files /dev/null and b/whisper-basic/test.wav differ
diff --git a/whisper-basic/whisper-basic.wasm b/whisper-basic/whisper-basic.wasm
new file mode 100755
index 0000000..7c55cd9
Binary files /dev/null and b/whisper-basic/whisper-basic.wasm differ