🔊 fix: Validate language format for OpenAI STT model (#10875)

* 🔊 fix: Validate language format for OpenAI STT model

* fix: Normalize input language model assignment in STTService

* refactor: Enhance error logging and language validation in STT and TTS services

* fix: Improve language validation in getValidatedLanguageCode function

api/server/services/Files/Audio/STTService.js

@@ -3,8 +3,8 @@ const fs = require('fs').promises;
 const FormData = require('form-data');
 const { Readable } = require('stream');
 const { logger } = require('@librechat/data-schemas');
-const { genAzureEndpoint } = require('@librechat/api');
 const { HttpsProxyAgent } = require('https-proxy-agent');
+const { genAzureEndpoint, logAxiosError } = require('@librechat/api');
 const { extractEnvVariable, STTProviders } = require('librechat-data-provider');
 const { getAppConfig } = require('~/server/services/Config');
 
@@ -35,6 +35,34 @@ const MIME_TO_EXTENSION_MAP = {
   'audio/x-flac': 'flac',
 };
 
+/**
+ * Validates and extracts ISO-639-1 language code from a locale string.
+ * @param {string} language - The language/locale string (e.g., "en-US", "en", "zh-CN")
+ * @returns {string|null} The ISO-639-1 language code (e.g., "en") or null if invalid
+ */
+function getValidatedLanguageCode(language) {
+  try {
+    if (!language) {
+      return null;
+    }
+
+    const normalizedLanguage = language.toLowerCase();
+    const isValidLocaleCode = /^[a-z]{2}(-[a-z]{2})?$/.test(normalizedLanguage);
+
+    if (isValidLocaleCode) {
+      return normalizedLanguage.split('-')[0];
+    }
+
+    logger.warn(
+      `[STT] Invalid language format "${language}". Expected ISO-639-1 locale code like "en-US" or "en". Skipping language parameter.`,
+    );
+    return null;
+  } catch (error) {
+    logger.error(`[STT] Error validating language code "${language}":`, error);
+    return null;
+  }
+}
+
 /**
  * Gets the file extension from the MIME type.
  * @param {string} mimeType - The MIME type.
@@ -173,10 +201,9 @@ class STTService {
       model: sttSchema.model,
     };
 
-    if (language) {
+    const validLanguage = getValidatedLanguageCode(language);
-      /** Converted locale code (e.g., "en-US") to ISO-639-1 format (e.g., "en") */
+    if (validLanguage) {
-      const isoLanguage = language.split('-')[0];
+      data.language = validLanguage;
-      data.language = isoLanguage;
     }
 
     const headers = {
@@ -221,10 +248,9 @@ class STTService {
       contentType: audioFile.mimetype,
     });
 
-    if (language) {
+    const validLanguage = getValidatedLanguageCode(language);
-      /** Converted locale code (e.g., "en-US") to ISO-639-1 format (e.g., "en") */
+    if (validLanguage) {
-      const isoLanguage = language.split('-')[0];
+      formData.append('language', validLanguage);
-      formData.append('language', isoLanguage);
     }
 
     const headers = {
@@ -286,7 +312,7 @@ class STTService {
 
       return response.data.text.trim();
     } catch (error) {
-      logger.error(`STT request failed for provider ${provider}:`, error);
+      logAxiosError({ message: `STT request failed for provider ${provider}:`, error });
       throw error;
     }
   }
@@ -316,7 +342,7 @@ class STTService {
       const text = await this.sttRequest(provider, sttSchema, { audioBuffer, audioFile, language });
       res.json({ text });
     } catch (error) {
-      logger.error('An error occurred while processing the audio:', error);
+      logAxiosError({ message: 'An error occurred while processing the audio:', error });
       res.sendStatus(500);
     } finally {
       try {

api/server/services/Files/Audio/TTSService.js

@@ -1,7 +1,7 @@
 const axios = require('axios');
 const { logger } = require('@librechat/data-schemas');
-const { genAzureEndpoint } = require('@librechat/api');
 const { HttpsProxyAgent } = require('https-proxy-agent');
+const { genAzureEndpoint, logAxiosError } = require('@librechat/api');
 const { extractEnvVariable, TTSProviders } = require('librechat-data-provider');
 const { getRandomVoiceId, createChunkProcessor, splitTextIntoChunks } = require('./streamAudio');
 const { getAppConfig } = require('~/server/services/Config');
@@ -274,7 +274,7 @@ class TTSService {
     try {
       return await axios.post(url, data, options);
     } catch (error) {
-      logger.error(`TTS request failed for provider ${provider}:`, error);
+      logAxiosError({ message: `TTS request failed for provider ${provider}:`, error });
       throw error;
     }
   }
@@ -330,7 +330,10 @@ class TTSService {
             break;
           }
         } catch (innerError) {
-          logger.error('Error processing manual update:', chunk, innerError);
+          logAxiosError({
+            message: `[TTS] Error processing manual update for chunk: ${chunk?.text?.substring(0, 50)}...`,
+            error: innerError,
+          });
           if (!res.headersSent) {
             return res.status(500).end();
           }
@@ -342,7 +345,7 @@ class TTSService {
         res.end();
       }
     } catch (error) {
-      logger.error('Error creating the audio stream:', error);
+      logAxiosError({ message: '[TTS] Error creating the audio stream:', error });
       if (!res.headersSent) {
         return res.status(500).send('An error occurred');
       }
@@ -412,7 +415,10 @@ class TTSService {
               break;
             }
           } catch (innerError) {
-            logger.error('Error processing audio stream update:', update, innerError);
+            logAxiosError({
+              message: `[TTS] Error processing audio stream update: ${update?.text?.substring(0, 50)}...`,
+              error: innerError,
+            });
             if (!res.headersSent) {
               return res.status(500).end();
             }
@@ -429,7 +435,7 @@ class TTSService {
         res.end();
       }
     } catch (error) {
-      logger.error('Failed to fetch audio:', error);
+      logAxiosError({ message: '[TTS] Failed to fetch audio:', error });
       if (!res.headersSent) {
         res.status(500).end();
       }