diff --git a/backend/open_webui/config.py b/backend/open_webui/config.py
index e735ea9a72..f89738b0db 100644
--- a/backend/open_webui/config.py
+++ b/backend/open_webui/config.py
@@ -1532,6 +1532,8 @@ AUDIO_STT_OPENAI_API_BASE_URL = os.getenv('AUDIO_STT_OPENAI_API_BASE_URL', OPENA
 
 AUDIO_STT_OPENAI_API_KEY = os.getenv('AUDIO_STT_OPENAI_API_KEY', OPENAI_API_KEY)
 
+AUDIO_STT_OPENAI_API_REQUEST_FORMAT = os.getenv('AUDIO_STT_OPENAI_API_REQUEST_FORMAT', 'multipart')
+
 AUDIO_STT_ENGINE = os.getenv('AUDIO_STT_ENGINE', '')
 
 AUDIO_STT_MODEL = os.getenv('AUDIO_STT_MODEL', '')
@@ -2966,6 +2968,7 @@ DEFAULT_CONFIG = {
     'audio.stt.deepgram.api_key': DEEPGRAM_API_KEY,
     'audio.stt.openai.api_base_url': AUDIO_STT_OPENAI_API_BASE_URL,
     'audio.stt.openai.api_key': AUDIO_STT_OPENAI_API_KEY,
+    'audio.stt.openai.api_request_format': AUDIO_STT_OPENAI_API_REQUEST_FORMAT,
     'audio.stt.engine': AUDIO_STT_ENGINE,
     'audio.stt.model': AUDIO_STT_MODEL,
     'audio.stt.supported_content_types': AUDIO_STT_SUPPORTED_CONTENT_TYPES,
diff --git a/backend/open_webui/routers/audio.py b/backend/open_webui/routers/audio.py
index 8310f49a2c..f062b51a37 100644
--- a/backend/open_webui/routers/audio.py
+++ b/backend/open_webui/routers/audio.py
@@ -94,6 +94,7 @@ TTS_CONFIG_KEYS = {
 STT_CONFIG_KEYS = {
     'OPENAI_API_BASE_URL': 'audio.stt.openai.api_base_url',
     'OPENAI_API_KEY': 'audio.stt.openai.api_key',
+    'OPENAI_API_REQUEST_FORMAT': 'audio.stt.openai.api_request_format',
     'ENGINE': 'audio.stt.engine',
     'MODEL': 'audio.stt.model',
     'SUPPORTED_CONTENT_TYPES': 'audio.stt.supported_content_types',
@@ -252,6 +253,7 @@ class TTSConfigForm(BaseModel):
 class STTConfigForm(BaseModel):
     OPENAI_API_BASE_URL: str
     OPENAI_API_KEY: str
+    OPENAI_API_REQUEST_FORMAT: str = 'multipart'
     ENGINE: str
     MODEL: str
     SUPPORTED_CONTENT_TYPES: list[str] = []
@@ -644,28 +646,49 @@ async def _transcribe_openai(request, file_path, filename, languages, file_dir,
     r = None
     try:
         session = await get_session()
+        api_key = await Config.get('audio.stt.openai.api_key')
+        api_base_url = await Config.get('audio.stt.openai.api_base_url')
+        request_format = (
+            await Config.get('audio.stt.openai.api_request_format') or 'multipart'
+        ).lower()
+
+        headers = {'Authorization': f'Bearer {api_key}'}
+        if user and ENABLE_FORWARD_USER_INFO_HEADERS:
+            headers = include_user_info_headers(headers, user)
+
         for language in languages:
             payload = {'model': await Config.get('audio.stt.model')}
             if language:
                 payload['language'] = language
-            api_key = await Config.get('audio.stt.openai.api_key')
-            api_base_url = await Config.get('audio.stt.openai.api_base_url')
 
-            headers = {'Authorization': f'Bearer {api_key}'}
-            if user and ENABLE_FORWARD_USER_INFO_HEADERS:
-                headers = include_user_info_headers(headers, user)
+            if request_format == 'json':
+                ext = os.path.splitext(filename)[1].lower().lstrip('.') or 'wav'
+                async with aiofiles.open(file_path, 'rb') as f:
+                    payload['input_audio'] = {
+                        'data': base64.b64encode(await f.read()).decode('utf-8'),
+                        'format': 'ogg' if ext == 'oga' else ext,
+                    }
 
-            form_data = aiohttp.FormData()
-            for key, value in payload.items():
-                form_data.add_field(key, str(value))
-            form_data.add_field('file', open(file_path, 'rb'), filename=filename)
+                r = await session.post(
+                    url=f'{api_base_url}/audio/transcriptions',
+                    headers={**headers, 'Content-Type': 'application/json'},
+                    json=payload,
+                    ssl=AIOHTTP_CLIENT_SESSION_SSL,
+                )
+            else:
+                form_data = aiohttp.FormData()
+                for key, value in payload.items():
+                    form_data.add_field(key, str(value))
 
-            r = await session.post(
-                url=f'{api_base_url}/audio/transcriptions',
-                headers=headers,
-                data=form_data,
-                ssl=AIOHTTP_CLIENT_SESSION_SSL,
-            )
+                with open(file_path, 'rb') as audio_file:
+                    form_data.add_field('file', audio_file, filename=filename)
+
+                    r = await session.post(
+                        url=f'{api_base_url}/audio/transcriptions',
+                        headers=headers,
+                        data=form_data,
+                        ssl=AIOHTTP_CLIENT_SESSION_SSL,
+                    )
             if r.status == 200:
                 break
 
diff --git a/src/lib/components/admin/Settings/Audio.svelte b/src/lib/components/admin/Settings/Audio.svelte
index cf893a24b0..a9a7fad001 100644
--- a/src/lib/components/admin/Settings/Audio.svelte
+++ b/src/lib/components/admin/Settings/Audio.svelte
@@ -43,6 +43,7 @@
 
 	let STT_OPENAI_API_BASE_URL = '';
 	let STT_OPENAI_API_KEY = '';
+	let STT_OPENAI_API_REQUEST_FORMAT = 'multipart';
 	let STT_ENGINE = '';
 	let STT_MODEL = '';
 	let STT_SUPPORTED_CONTENT_TYPES = '';
@@ -150,6 +151,7 @@
 			stt: {
 				OPENAI_API_BASE_URL: STT_OPENAI_API_BASE_URL,
 				OPENAI_API_KEY: STT_OPENAI_API_KEY,
+				OPENAI_API_REQUEST_FORMAT: STT_OPENAI_API_REQUEST_FORMAT,
 				ENGINE: STT_ENGINE,
 				MODEL: STT_MODEL,
 				SUPPORTED_CONTENT_TYPES: STT_SUPPORTED_CONTENT_TYPES.split(','),
@@ -202,6 +204,7 @@
 
 			STT_OPENAI_API_BASE_URL = res.stt.OPENAI_API_BASE_URL;
 			STT_OPENAI_API_KEY = res.stt.OPENAI_API_KEY;
+			STT_OPENAI_API_REQUEST_FORMAT = res.stt.OPENAI_API_REQUEST_FORMAT || 'multipart';
 
 			STT_ENGINE = res.stt.ENGINE;
 			STT_MODEL = res.stt.MODEL;
@@ -288,6 +291,21 @@
 
 					<hr class="border-gray-100/30 dark:border-gray-850/30 my-2" />
 
+					<div class="mb-2 py-0.5 flex w-full justify-between">
+						<div class=" self-center text-xs font-medium">{$i18n.t('Request Format')}</div>
+						<div class="flex items-center relative">
+							<select
+								class="cursor-pointer w-fit pr-8 rounded-sm px-2 p-1 text-xs bg-transparent outline-hidden text-right"
+								bind:value={STT_OPENAI_API_REQUEST_FORMAT}
+							>
+								<option value="multipart">{$i18n.t('Multipart Upload')}</option>
+								<option value="json">{$i18n.t('JSON Base64')}</option>
+							</select>
+						</div>
+					</div>
+
+					<hr class="border-gray-100/30 dark:border-gray-850/30 my-2" />
+
 					<div>
 						<div class=" mb-1.5 text-xs font-medium">{$i18n.t('STT Model')}</div>
 						<div class="flex w-full">