diff --git a/yt_dlp/downloader/twitter.py b/yt_dlp/downloader/twitter.py
index d01df98a1b..fa706e3140 100644
--- a/yt_dlp/downloader/twitter.py
+++ b/yt_dlp/downloader/twitter.py
@@ -9,7 +9,7 @@ def _read_fragment(self, ctx):
         idx = frag_content.find(bytes(
             [0x49, 0x44, 0x33, 0x04, 0x00, 0x00, 0x00, 0x00, 0x00, 0x3F, 0x50, 0x52, 0x49, 0x56]))
         if idx > 0:
-            self.to_screen(f'[{self.FD_NAME}] Removing partial header ({idx} bytes)')
+            # self.to_screen(f'[{self.FD_NAME}] Removing partial header ({idx} bytes)')
             return frag_content[idx:]
         else:
             return frag_content
diff --git a/yt_dlp/extractor/twitter.py b/yt_dlp/extractor/twitter.py
index dae0780183..a20ece6f19 100644
--- a/yt_dlp/extractor/twitter.py
+++ b/yt_dlp/extractor/twitter.py
@@ -1852,7 +1852,7 @@ def _real_extract(self, url):
                 ('source', ('noRedirectPlaybackUrl', 'location'), {url_or_none}), get_all=False)
             is_audio_space = source and 'audio-space' in source
             formats = self._extract_m3u8_formats(
-                source, metadata['media_key'], 'm4a' if is_audio_space else 'mp4',
+                source, metadata['media_key'], 'aac' if is_audio_space else 'mp4',
                 # XXX: Some audio-only Spaces need to truncate heads of their fragments, not only ffmpeg as downloader
                 entry_protocol='twitter_spaces' if is_audio_space else 'm3u8_native',
                 live=is_live, headers=headers, fatal=False) if source else []