diff --git a/fern/docs.yml b/fern/docs.yml
index 77dc49b7..ea5568f7 100644
--- a/fern/docs.yml
+++ b/fern/docs.yml
@@ -1454,9 +1454,9 @@ redirects:
   - source: /docs/speech-to-text/pre-recorded-audio/ald-99-early-access
     destination: /docs/speech-to-text/pre-recorded-audio/automatic-language-detection
   - source: /docs/speech-to-text/pre-recorded-audio/keyterms-prompting-slam-1
-    destination: /docs/pre-recorded-audio/improving-transcript-results-with-keyterms-prompting
-  - source: /docs/pre-recorded-audio/improving-transcript-accuracy
-    destination: /docs/pre-recorded-audio/improving-transcript-results-with-keyterms-prompting
+    destination: /docs/pre-recorded-audio/improving-transcript-accuracy
+  - source: /docs/pre-recorded-audio/improving-transcript-results-with-keyterms-prompting
+    destination: /docs/pre-recorded-audio/improving-transcript-accuracy
   - source: /docs/faq/does-assemblyai-offer-zero-data-retention
     destination: /docs/faq/how-long-does-aai-retain-data
   - source: /docs/security-compliance
@@ -1501,6 +1501,15 @@ redirects:
   #   destination: /docs/llm-gateway/overview
 
   # Redirects for Speech to Text
+  # Specific redirects must come before wildcard redirects
+  - source: /docs/speech-to-text/pre-recorded-audio/set-language-manually
+    destination: /docs/pre-recorded-audio/language-detection#set-language-manually
+  - source: /docs/speech-to-text/pre-recorded-audio/word-level-timestamps
+    destination: /docs/pre-recorded-audio/transcript-export-options#word-level-timestamps
+  - source: /docs/speech-to-text/pre-recorded-audio/improving-transcript-accuracy
+    destination: /docs/pre-recorded-audio/keyterms-prompting
+  - source: /docs/speech-to-text/pre-recorded-audio/automatic-language-detection
+    destination: /docs/pre-recorded-audio/language-detection
   - source: /docs/speech-to-text/pre-recorded-audio/:slug
     destination: /docs/pre-recorded-audio/:slug
   - source: /docs/speech-to-text/universal-streaming/:slug
@@ -1561,19 +1570,13 @@ redirects:
     destination: /docs/getting-started/universal-3-pro
   - source: /docs/pre-recorded-audio/prompt-engineering
     destination: /docs/pre-recorded-audio/prompting
-  - source: /docs/speech-to-text/pre-recorded-audio/improving-transcript-accuracy
-    destination: /docs/pre-recorded-audio/keyterms-prompting
-  - source: /docs/speech-to-text/pre-recorded-audio/automatic-language-detection
-    destination: /docs/pre-recorded-audio/language-detection
   - source: /docs/pre-recorded-audio/automatic-language-detection
     destination: /docs/pre-recorded-audio/language-detection
   - source: /docs/pre-recorded-audio/improving-transcript-accuracy
     destination: /docs/pre-recorded-audio/keyterms-prompting
   - source: /playground/source
     destination: https://www.assemblyai.com/dashboard/playground
-  - source: /docs/speech-to-text/pre-recorded-audio/word-level-timestamps
-    destination: /docs/pre-recorded-audio/transcript-export-options
   - source: /docs/pre-recorded-audio/export-paragraphs-and-sentences
     destination: /docs/pre-recorded-audio/transcript-export-options
-  - source: /docs/speech-to-text/pre-recorded-audio/set-language-manually
-    destination: https://www.assemblyai.com/docs/pre-recorded-audio/language-detection#set-language-manually
+  - source: /docs/pre-recorded-audio/improving-transcript-results-with-keyterms-prompting
+    destination: /docs/pre-recorded-audio/improving-transcript-accuracy