diff --git a/demos/audio/README.md b/demos/audio/README.md
index 6500913baa..27f0d5b73e 100644
--- a/demos/audio/README.md
+++ b/demos/audio/README.md
@@ -24,6 +24,7 @@ mkdir -p audio_samples
 curl --output audio_samples/audio.wav "https://www.voiptroubleshooter.com/open_speech/american/OSR_us_000_0032_8k.wav"
 mkdir -p models
 mkdir -p models/speakers
+curl https://raw.githubusercontent.com/openvinotoolkit/model_server/refs/heads/releases/2026/0/demos/audio/create_speaker_embedding.py -o create_speaker_embedding.py
 python create_speaker_embedding.py audio_samples/audio.wav models/speakers/voice1.bin
 ```
 
diff --git a/demos/common/export_models/export_model.py b/demos/common/export_models/export_model.py
index b7b6a55b72..5aa81b0c81 100644
--- a/demos/common/export_models/export_model.py
+++ b/demos/common/export_models/export_model.py
@@ -40,7 +40,7 @@ def add_common_arguments(parser):
 add_common_arguments(parser_text)
 parser_text.add_argument('--pipeline_type', default=None, choices=["LM", "LM_CB", "VLM", "VLM_CB", "AUTO"], help='Type of the pipeline to be used. AUTO is used by default', dest='pipeline_type')
 parser_text.add_argument('--kv_cache_precision', default=None, choices=["u8"], help='u8 or empty (model default). Reduced kv cache precision to u8 lowers the cache size consumption.', dest='kv_cache_precision')
-parser_text.add_argument('--enable_prefix_caching', type=lambda x: (str(x).lower() == 'true'), default=True, help='This algorithm is used to cache the prompt tokens.', dest='enable_prefix_caching')
+parser_text.add_argument('--enable_prefix_caching', type=lambda x: (str(x).lower() == 'true'), default=True, help='This algorithm is used to cache the prompt tokens. Default is True.', dest='enable_prefix_caching')
 parser_text.add_argument('--disable_dynamic_split_fuse', action='store_false', help='The maximum number of tokens that can be batched together.', dest='dynamic_split_fuse')
 parser_text.add_argument('--max_num_batched_tokens', default=None, help='empty or integer. The maximum number of tokens that can be batched together.', dest='max_num_batched_tokens')
 parser_text.add_argument('--max_num_seqs', default=None, help='256 by default. The maximum number of sequences that can be processed together.', dest='max_num_seqs')