diff --git a/pyopenagi/agents/om-raheja/transcribe_agent/agent.py b/pyopenagi/agents/om-raheja/transcribe_agent/agent.py
new file mode 100755
index 00000000..b9c3eb57
--- /dev/null
+++ b/pyopenagi/agents/om-raheja/transcribe_agent/agent.py
@@ -0,0 +1,17 @@
+from ...react_agent import ReactAgent
+
+class TranscribeAgent(ReactAgent):
+    def __init__(self,
+                 agent_name,
+                 task_input,
+                 agent_process_factory,
+                 log_mode: str
+        ):
+        ReactAgent.__init__(self, agent_name, task_input, agent_process_factory, log_mode)
+        self.workflow_mode = "automatic"
+
+    def manual_workflow(self):
+        pass
+
+    def run(self):
+        return super().run()
diff --git a/pyopenagi/agents/om-raheja/transcribe_agent/config.json b/pyopenagi/agents/om-raheja/transcribe_agent/config.json
new file mode 100644
index 00000000..93089ef3
--- /dev/null
+++ b/pyopenagi/agents/om-raheja/transcribe_agent/config.json
@@ -0,0 +1,14 @@
+{
+    "name": "transcribe_agent",
+    "description": [
+        "You are an agent who can transcribe audio from the microphone into text. "
+    ],
+    "tools": [
+        "transcriber/transcriber"
+    ],
+    "meta": {
+        "author": "Om Raheja",
+        "version": "0.0.1",
+        "license": "CC0"
+    }
+}
diff --git a/pyopenagi/agents/om-raheja/transcribe_agent/meta_requirements.txt b/pyopenagi/agents/om-raheja/transcribe_agent/meta_requirements.txt
new file mode 100644
index 00000000..d7d6bed4
--- /dev/null
+++ b/pyopenagi/agents/om-raheja/transcribe_agent/meta_requirements.txt
@@ -0,0 +1 @@
+RealtimeSTT
diff --git a/pyopenagi/tools/transcriber/transcriber.py b/pyopenagi/tools/transcriber/transcriber.py
new file mode 100644
index 00000000..ce5aa7aa
--- /dev/null
+++ b/pyopenagi/tools/transcriber/transcriber.py
@@ -0,0 +1,52 @@
+from ..base import BaseTool 
+from time import sleep
+
+class CurrencyConverter(BaseTool):
+    def __init__(self):
+        """ big library, not everyone needs it installed """
+        try:
+            from RealtimeSTT import AudioToTextRecorder
+        except ImportError:
+            raise ImportError(
+                "Please install RealtimeSTT: `pip install RealtimeSTT`"
+            )
+        
+        # this is hardcoded for now
+        self.recorder = AudioToTextRecorder(
+                model="tiny.en",
+        )
+
+    def run(self, params: dict):
+        duration = 5
+        try:
+            duration = int(params["duration"])
+        except ValueError:
+            raise KeyError(
+                "The keys in params do not match the excepted key in params for transcriber api. "
+                "Please make sure it contain the key 'duration'"
+            )
+
+        self.record.start()
+        sleep(duration)
+        self.recorder.stop()
+        return self.recorder.text() 
+
+    def get_tool_call_format(self):
+        tool_call_format = {
+            "type": "function",
+            "function": {
+                "name": "transcriber",
+                "description": "Transcribes audiio into text",
+                "parameters": {
+                    "type": "object",
+                    "properties": {
+                        "duration": {
+                            "type": "string",
+                            "description": "How long to record audio for in seconds",
+                        },
+                    },
+                    "required": []
+                }
+            }
+        }
+        return tool_call_format