diff --git a/README.md b/README.md
index 852be949..75080871 100644
--- a/README.md
+++ b/README.md
@@ -5,7 +5,7 @@
 </div>
 
 
-Hugging Face Inference Toolkit is for serving 🤗 Transformers models in containers. This library provides default pre-processing, predict and postprocessing for Transformers, Sentence Tranfsformers. It is also possible to define custom `pipeline.py` for customization. The Toolkit is build to work with the [Hugging Face Hub](https://huggingface.co/models).
+Hugging Face Inference Toolkit is for serving 🤗 Transformers models in containers. This library provides default pre-processing, predict and postprocessing for Transformers, Sentence Tranfsformers. It is also possible to define custom `handler.py` for customization. The Toolkit is build to work with the [Hugging Face Hub](https://huggingface.co/models).
 
 ---
 ## 💻  Getting Started with Hugging Face Inference Toolkit
@@ -124,24 +124,24 @@ HF_ENDPOINT="True"
 
 ---
 
-## 🧑🏻‍💻 Custom Pipeline and dependency support
+## 🧑🏻‍💻 Custom Handler and dependency support
 
-The Hugging Face Inference Toolkit allows user to provide a custom inference through a `pipeline.py` file which is located in the repository. 
+The Hugging Face Inference Toolkit allows user to provide a custom inference through a `handler.py` file which is located in the repository. 
 For an example check [https://huggingface.co/philschmid/custom-pipeline-text-classification](https://huggingface.co/philschmid/custom-pipeline-text-classification):  
 ```bash
 model.tar.gz/
 |- pytorch_model.bin
 |- ....
-|- pipeline.py
+|- handler.py
 |- requirements.txt 
 ```
-In this example, `pytroch_model.bin` is the model file saved from training, `pipeline.py` is the custom inference pipeline, and `requirements.txt` is a requirements file to add additional dependencies.
+In this example, `pytroch_model.bin` is the model file saved from training, `handler.py` is the custom inference handler, and `requirements.txt` is a requirements file to add additional dependencies.
 The custom module can override the following methods:  
 
 
 ## ☑️ Supported & Tested Tasks
 
-Below you ll find a list of supported and tested transformers and sentence transformers tasks. Each of those are always tested through integration tests. In addition to those tasks you can always provide `custom`, which expect a `pipeline.py` file to be provided.
+Below you ll find a list of supported and tested transformers and sentence transformers tasks. Each of those are always tested through integration tests. In addition to those tasks you can always provide `custom`, which expect a `handler.py` file to be provided.
 
 ```bash
 "text-classification",
diff --git a/src/huggingface_inference_toolkit/serialization/base.py b/src/huggingface_inference_toolkit/serialization/base.py
index 731139c4..38a4a5d8 100644
--- a/src/huggingface_inference_toolkit/serialization/base.py
+++ b/src/huggingface_inference_toolkit/serialization/base.py
@@ -25,6 +25,8 @@
     "audio/x-wav": Audioer,
     "audio/ogg": Audioer,
     "audio/x-audio": Audioer,
+    "audio/webm": Audioer,
+    "audio/webm;codecs=opus": Audioer,
 }
 
 
diff --git a/src/huggingface_inference_toolkit/webservice_starlette.py b/src/huggingface_inference_toolkit/webservice_starlette.py
index 933534b7..76ca5478 100644
--- a/src/huggingface_inference_toolkit/webservice_starlette.py
+++ b/src/huggingface_inference_toolkit/webservice_starlette.py
@@ -82,10 +82,10 @@ async def predict(request):
         # TODO: repalce with middleware
         logger.info(f"POST {request.url.path} | Duration: {(perf_counter()-start_time) *1000:.2f} ms")
         # deserialized and resonds with json
-        return Response(Jsoner.serialize(pred))
+        return Response(Jsoner.serialize(pred), media_type="application/json")
     except Exception as e:
         logger.error(e)
-        return Response(Jsoner.serialize({"error": str(e)}), status_code=400)
+        return Response(Jsoner.serialize({"error": str(e)}), status_code=400, media_type="application/json")
 
 
 app = Starlette(
diff --git a/tests/resources/audio/sample1.webm b/tests/resources/audio/sample1.webm
new file mode 100644
index 00000000..b74348cc
Binary files /dev/null and b/tests/resources/audio/sample1.webm differ
diff --git a/tests/unit/test_utils.py b/tests/unit/test_utils.py
index 7ed732ef..b35cb734 100644
--- a/tests/unit/test_utils.py
+++ b/tests/unit/test_utils.py
@@ -69,7 +69,7 @@ def test_load_onnx_repository_from_hf():
         # onnx model
         assert "model.onnx" in folder_contents
         # custom pipeline
-        assert "pipeline.py" in folder_contents
+        assert "handler.py" in folder_contents
         # revision doesn't have tokenizer
         assert "tokenizer_config.json" in folder_contents