Update audio transcriptions tests with a smaller, reliable audio file source (#351)

Author: yadavsahil197

tests/integration/resources/test_transcriptions.py

@@ -22,7 +22,9 @@ def test_basic_transcription_url(self, sync_together_client):
         """
         Test basic transcription with URL audio file
         """
-        audio_url = "https://ia801605.us.archive.org/28/items/jfks19630626/jfk_1963_0626_berliner.wav"
+        audio_url = (
+            "https://voiptroubleshooter.com/open_speech/american/OSR_us_000_0010_8k.wav"
+        )
 
         response = sync_together_client.audio.transcriptions.create(
             file=audio_url, model="openai/whisper-large-v3"
@@ -36,7 +38,9 @@ def test_transcription_with_language(self, sync_together_client):
         """
         Test transcription with language parameter
         """
-        audio_url = "https://ia801605.us.archive.org/28/items/jfks19630626/jfk_1963_0626_berliner.wav"
+        audio_url = (
+            "https://voiptroubleshooter.com/open_speech/american/OSR_us_000_0010_8k.wav"
+        )
 
         response = sync_together_client.audio.transcriptions.create(
             file=audio_url, model="openai/whisper-large-v3", language="en"
@@ -50,7 +54,9 @@ def test_transcription_verbose_json(self, sync_together_client):
         """
         Test transcription with verbose JSON format and timestamps
         """
-        audio_url = "https://ia801605.us.archive.org/28/items/jfks19630626/jfk_1963_0626_berliner.wav"
+        audio_url = (
+            "https://voiptroubleshooter.com/open_speech/american/OSR_us_000_0010_8k.wav"
+        )
 
         response = sync_together_client.audio.transcriptions.create(
             file=audio_url,
@@ -68,7 +74,9 @@ def test_transcription_with_temperature(self, sync_together_client):
         """
         Test transcription with temperature parameter
         """
-        audio_url = "https://ia801605.us.archive.org/28/items/jfks19630626/jfk_1963_0626_berliner.wav"
+        audio_url = (
+            "https://voiptroubleshooter.com/open_speech/american/OSR_us_000_0010_8k.wav"
+        )
 
         response = sync_together_client.audio.transcriptions.create(
             file=audio_url, model="openai/whisper-large-v3", temperature=0.2
@@ -91,7 +99,9 @@ def test_transcription_missing_model(self, sync_together_client):
         """
         Test transcription with missing model parameter - should use default model
         """
-        audio_url = "https://ia801605.us.archive.org/28/items/jfks19630626/jfk_1963_0626_berliner.wav"
+        audio_url = (
+            "https://voiptroubleshooter.com/open_speech/american/OSR_us_000_0010_8k.wav"
+        )
 
         response = sync_together_client.audio.transcriptions.create(file=audio_url)