update eg

Author: TudorCRL

examples/rt/async/speaker_id/README.md

@@ -8,3 +8,8 @@ The SDK requires an API key to be set as an environment variable before it can b
 
 - Install Speechmatics RT SDK: `pip install speechmatics-rt`
 - Export Speechmatics API key: `export SPEECHMATICS_API_KEY=YOUR-API-KEY`
+
+## Usage
+
+- Generate speaker IDs: `python generate.py` - this will generate a `speakers.json` file
+- Transcribe audio: `python transcribe.py` - this will use the `speakers.json` file to perform speaker ID on a conversation

examples/rt/async/speaker_id/generate.py

@@ -0,0 +1,56 @@
+import asyncio
+import logging
+import os
+import json
+
+from speechmatics.rt import (
+    AsyncClient,
+    OperatingPoint,
+    TranscriptionConfig,
+    ServerMessageType,
+)
+
+
+logging.basicConfig(level=logging.INFO)
+
+
+speakers: list[dict] = []
+
+
+async def generate_ids(voice_file: str) -> None:
+    """Run async transcription example."""
+
+    transcription_config = TranscriptionConfig(
+        operating_point=OperatingPoint.ENHANCED,
+        diarization="speaker",
+    )
+
+    # Initialize client with API key from environment
+    async with AsyncClient() as client:
+        try:
+            @client.on(ServerMessageType.SPEAKERS_RESULT)
+            def handle_speakers_result(msg):
+                new_speakers = msg.get('speakers', [])
+                new_speakers[0]["label"] = voice_file
+                speakers.append(new_speakers[0])
+
+            # Transcribe audio file
+            with open(os.path.join(voices_folder, voice_file), "rb") as audio_file:
+                await client.transcribe(
+                    audio_file,
+                    transcription_config=transcription_config,
+                    get_speakers=True,
+                )
+        except Exception as e:
+            print(f"Transcription error: {e}")
+
+
+if __name__ == "__main__":
+    voices_folder = "./examples/rt/async/speaker_id/voices"
+    voice_files = [f for f in os.listdir(voices_folder) if os.path.isfile(os.path.join(voices_folder, f))]
+
+    for voice_file in voice_files:
+        asyncio.run(generate_ids(voice_file))
+
+    with open('./speakers.json', 'w') as f:
+        json.dump(speakers, f)

examples/rt/async/speaker_id/main.py renamed to examples/rt/async/speaker_id/transcribe.py

@@ -1,13 +1,14 @@
 import asyncio
 import logging
+import json
 
-from speechmatics.rt import ServerMessageType
+from speechmatics.rt import SpeakerIdentifier
+from speechmatics.rt import SpeakerDiarizationConfig
 from speechmatics.rt import (
     AsyncClient,
-    AudioEncoding,
-    AudioFormat,
     OperatingPoint,
     TranscriptionConfig,
+    ServerMessageType
 )
 
 
@@ -17,29 +18,30 @@
 async def main() -> None:
     """Run async transcription example."""
 
+    with open('./speakers.json') as f:
+        speaker_identifiers = [SpeakerIdentifier(**s) for s in json.load(f)]
+
     transcription_config = TranscriptionConfig(
-        max_delay=0.8,
         operating_point=OperatingPoint.ENHANCED,
         diarization="speaker",
+        max_delay=4,
+        speaker_diarization_config=SpeakerDiarizationConfig(
+            speakers=speaker_identifiers,
+        )
     )
 
     # Initialize client with API key from environment
     async with AsyncClient() as client:
         try:
             @client.on(ServerMessageType.ADD_TRANSCRIPT)
             def handle_finals(msg):
-                print(f"Final: {msg['metadata']['transcript']}")
-
-            @client.on(ServerMessageType.SPEAKERS_RESULT)
-            def handle_speakers_result(msg):
-                print(msg)
+                print(f"Final: {msg['metadata']['speaker']} {msg['metadata']['transcript']}")
 
             # Transcribe audio file
-            with open("./examples/example.wav", "rb") as audio_file:
+            with open("./examples/conversation.wav", "rb") as audio_file:
                 await client.transcribe(
                     audio_file,
                     transcription_config=transcription_config,
-                    get_speakers=True,
                 )
         except Exception as e:
             print(f"Transcription error: {e}")