New speech models update for multiple locales

es-SV/es-EC/es-BO/es-PY/es-AR/es-DO/es-UY/es-CR/es-VE/es-NI/es-HN/es-PR/es-CO/es-CL/es-CU/es-PE/es-PA/es-GT/es-GQ
and 
pl-PL/zh-TW/id-ID/fi-FI/pl-OL

Author: jackykang7777

articles/ai-services/speech-service/includes/release-notes/release-notes-stt.md

@@ -8,6 +8,11 @@ ms.author: eur
 
 ### September 2024 release
 
+
+#### [Real-time speech to text](../../how-to-recognize-speech.md) has released new models for follwoing languages with better quality. 
+fi-FI/id-ID/zh-TW/pl-PL/pt-PT 
+es-SV/es-EC/es-BO/es-PY/es-AR/es-DO/es-UY/es-CR/es-VE/es-NI/es-HN/es-PR/es-CO/es-CL/es-CU/es-PE/es-PA/es-GT/es-GQ
+
 #### Fast transcription (Preview)
 Fast transcription now supports diarization to recognize and separate multiple speakers on mono channel audio file. For more information, see [fast transcription API guide](../../fast-transcription-create.md#use-the-fast-transcription-api).