From 76b9bbf473fe8909fb03ba585aa887cba4b47a74 Mon Sep 17 00:00:00 2001
From: duyalei <>
Date: Mon, 23 Sep 2024 15:15:54 +0800
Subject: [PATCH] azure stt: auto detect multi-languages

---
 src/pipecat/services/azure.py | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/src/pipecat/services/azure.py b/src/pipecat/services/azure.py
index 36c8bc1bb..72e62c43e 100644
--- a/src/pipecat/services/azure.py
+++ b/src/pipecat/services/azure.py
@@ -39,6 +39,7 @@
         SpeechSynthesizer,
         ResultReason,
         CancellationReason,
+        languageconfig
     )
     from azure.cognitiveservices.speech.audio import AudioStreamFormat, PushAudioInputStream
     from azure.cognitiveservices.speech.dialog import AudioConfig
@@ -139,14 +140,23 @@ def __init__(
         super().__init__(**kwargs)
 
         speech_config = SpeechConfig(subscription=api_key, region=region)
-        speech_config.speech_recognition_language = language
-
         stream_format = AudioStreamFormat(samples_per_second=sample_rate, channels=channels)
         self._audio_stream = PushAudioInputStream(stream_format)
 
         audio_config = AudioConfig(stream=self._audio_stream)
-        self._speech_recognizer = SpeechRecognizer(
-            speech_config=speech_config, audio_config=audio_config)
+        # supported languages are listed here:
+        # https://learn.microsoft.com/en-us/azure/ai-services/speech-service/language-support?tabs=language-identification
+        if isinstance(language, list):
+            self._speech_recognizer = SpeechRecognizer(
+                speech_config=speech_config,
+                audio_config=audio_config,
+                auto_detect_source_language_config=languageconfig.AutoDetectSourceLanguageConfig(
+                    languages=language)
+            )
+        else:
+            self._speech_recognizer = SpeechRecognizer(speech_config=speech_config,
+                                                       audio_config=audio_config,
+                                                       language=language)
         self._speech_recognizer.recognized.connect(self._on_handle_recognized)
 
     async def run_stt(self, audio: bytes) -> AsyncGenerator[Frame, None]: