diff --git a/src/bindings/python/tests/test_inference_engine/test_compiled_model.py b/src/bindings/python/tests/test_inference_engine/test_compiled_model.py
index 60b5f44e8fc..e0bb872bf4d 100644
--- a/src/bindings/python/tests/test_inference_engine/test_compiled_model.py
+++ b/src/bindings/python/tests/test_inference_engine/test_compiled_model.py
@@ -29,8 +29,8 @@ def test_get_property(device):
         pytest.skip("Can't run on ARM plugin due-to CPU dependent test")
     func = core.read_model(model=test_net_xml, weights=test_net_bin)
     exec_net = core.compile_model(func, device)
-    config = exec_net.get_property("PERF_COUNT")
-    assert config == "NO"
+    profiling_enabled = exec_net.get_property("PERF_COUNT")
+    assert not profiling_enabled
 
 
 def test_get_runtime_model(device):
diff --git a/src/plugins/intel_cpu/src/exec_network.cpp b/src/plugins/intel_cpu/src/exec_network.cpp
index b2f20cd9020..5116c57989a 100644
--- a/src/plugins/intel_cpu/src/exec_network.cpp
+++ b/src/plugins/intel_cpu/src/exec_network.cpp
@@ -26,6 +26,7 @@
 #include "cpp_interfaces/interface/ie_iplugin_internal.hpp"
 #include "ie_icore.hpp"
 #include "openvino/runtime/properties.hpp"
+#include "openvino/util/common_util.hpp"
 
 #include <algorithm>
 #include <unordered_set>
@@ -310,7 +311,8 @@ InferenceEngine::Parameter MKLDNNExecNetwork::GetMetric(const std::string &name)
 
     if (name == ov::model_name) {
         // @todo Does not seem ok to 'dump()' the whole graph everytime in order to get a name
-        return graph.dump()->get_friendly_name();
+        const std::string modelName = graph.dump()->get_friendly_name();
+        return decltype(ov::model_name)::value_type(modelName);
     } else if (name == ov::optimal_number_of_infer_requests) {
         const auto streams = config.streamExecutorConfig._streams;
         return decltype(ov::optimal_number_of_infer_requests)::value_type(streams); // ov::optimal_number_of_infer_requests has no negative values
@@ -332,19 +334,20 @@ InferenceEngine::Parameter MKLDNNExecNetwork::GetMetric(const std::string &name)
         return ov::Affinity::NONE;
     } else if (name == ov::inference_num_threads) {
         const auto num_threads = config.streamExecutorConfig._threads;
-        return num_threads;
+        return decltype(ov::inference_num_threads)::value_type(num_threads);
     } else if (name == ov::enable_profiling.name()) {
         const bool perfCount = config.collectPerfCounters;
-        return perfCount ? "YES" : "NO";
+        return decltype(ov::enable_profiling)::value_type(perfCount);
     } else if (name == ov::hint::inference_precision) {
         const auto enforceBF16 = config.enforceBF16;
-        return enforceBF16 ? ov::element::bf16 : ov::element::f32;
+        const auto inference_precision = enforceBF16 ? ov::element::bf16 : ov::element::f32;
+        return decltype(ov::hint::inference_precision)::value_type(inference_precision);
     } else if (name == ov::hint::performance_mode) {
-        const auto perfHint = config.perfHintsConfig.ovPerfHint;
+        const auto perfHint = ov::util::from_string(config.perfHintsConfig.ovPerfHint, ov::hint::performance_mode);
         return perfHint;
     } else if (name == ov::hint::num_requests) {
         const auto perfHintNumRequests = config.perfHintsConfig.ovPerfHintNumRequests;
-        return perfHintNumRequests;
+        return decltype(ov::hint::num_requests)::value_type(perfHintNumRequests);
     }
     /* Internally legacy parameters are used with new API as part of migration procedure.
      * This fallback can be removed as soon as migration completed */
diff --git a/src/plugins/intel_cpu/src/plugin.cpp b/src/plugins/intel_cpu/src/plugin.cpp
index b1e46577633..afe9058d4a7 100644
--- a/src/plugins/intel_cpu/src/plugin.cpp
+++ b/src/plugins/intel_cpu/src/plugin.cpp
@@ -106,6 +106,7 @@
 #include <low_precision/multiply_to_group_convolution.hpp>
 #include <low_precision/network_helper.hpp>
 #include "openvino/runtime/core.hpp"
+#include "openvino/util/common_util.hpp"
 
 #include <ie_algorithm.hpp>
 #include "performance_heuristics.hpp"
@@ -568,6 +569,7 @@ void Engine::ApplyPerformanceHints(std::map<std::string, std::string> &config, c
 
     if (mode_name == CONFIG_VALUE(LATENCY)) {
         config[CONFIG_KEY(CPU_THROUGHPUT_STREAMS)] = CONFIG_VALUE(CPU_THROUGHPUT_NUMA);
+        config[ov::num_streams.name()] = ov::util::to_string(ov::NumStreams(ov::NumStreams::NUMA));
     } else if (mode_name == CONFIG_VALUE(THROUGHPUT)) {
         const auto isa = dnnl::get_effective_cpu_isa();
         float isaSpecificThreshold = 1.0f;
@@ -625,6 +627,7 @@ void Engine::ApplyPerformanceHints(std::map<std::string, std::string> &config, c
                                    engConfig.perfHintsConfig.ovPerfHintNumRequests);
         }
         config[CONFIG_KEY(CPU_THROUGHPUT_STREAMS)] = std::to_string(num_streams);
+        config[ov::num_streams.name()] = ov::util::to_string(ov::NumStreams(num_streams));
     }
 }
 
@@ -745,11 +748,11 @@ Parameter Engine::GetConfig(const std::string& name, const std::map<std::string,
         return decltype(ov::enable_profiling)::value_type(perfCount);
     } else if (name == ov::hint::inference_precision) {
         const auto enforceBF16 = engConfig.enforceBF16;
-        return decltype(ov::hint::inference_precision)::value_type(
-            enforceBF16 ? ov::element::bf16 : ov::element::f32);
+        const auto inference_precision = enforceBF16 ? ov::element::bf16 : ov::element::f32;
+        return decltype(ov::hint::inference_precision)::value_type(inference_precision);
     } else if (name == ov::hint::performance_mode) {
-        const auto perfHint = engConfig.perfHintsConfig.ovPerfHint;
-        return ov::Any{perfHint}.as<decltype(ov::hint::performance_mode)::value_type>();
+        const auto perfHint = ov::util::from_string(engConfig.perfHintsConfig.ovPerfHint, ov::hint::performance_mode);
+        return perfHint;
     } else if (name == ov::hint::num_requests) {
         const auto perfHintNumRequests = engConfig.perfHintsConfig.ovPerfHintNumRequests;
         return decltype(ov::hint::num_requests)::value_type(perfHintNumRequests);
@@ -840,12 +843,13 @@ Parameter Engine::GetMetric(const std::string& name, const std::map<std::string,
         supportedProperties.reserve(roProperties.size() + rwProperties.size());
         supportedProperties.insert(supportedProperties.end(), roProperties.begin(), roProperties.end());
         supportedProperties.insert(supportedProperties.end(), rwProperties.begin(), rwProperties.end());
-        return supportedProperties;
+
+        return decltype(ov::supported_properties)::value_type(supportedProperties);
     } else if (name == ov::device::full_name) {
-        return deviceFullName;
+        return decltype(ov::device::full_name)::value_type(deviceFullName);
     } else if (name == ov::available_devices) {
         const std::vector<std::string> availableDevices = { "" };
-        return availableDevices;
+        return decltype(ov::available_devices)::value_type(availableDevices);
     } else if (name == ov::device::capabilities) {
         std::vector<std::string> capabilities;
         if (dnnl::impl::cpu::x64::mayiuse(dnnl::impl::cpu::x64::avx512_core_bf16))
@@ -857,13 +861,13 @@ Parameter Engine::GetMetric(const std::string& name, const std::map<std::string,
         capabilities.push_back(METRIC_VALUE(INT8));
         capabilities.push_back(METRIC_VALUE(BIN));
         capabilities.push_back(ov::device::capability::EXPORT_IMPORT);
-        return capabilities;
+        return decltype(ov::device::capabilities)::value_type(capabilities);
     } else if (name == ov::range_for_async_infer_requests) {
         const std::tuple<unsigned int, unsigned int, unsigned int> range = std::make_tuple(1, 1, 1);
-        return range;
+        return decltype(ov::range_for_async_infer_requests)::value_type(range);
     } else if (name == ov::range_for_streams) {
         const std::tuple<unsigned int, unsigned int> range = std::make_tuple(1, parallel_get_max_threads());
-        return range;
+        return decltype(ov::range_for_streams)::value_type(range);
     }
     /* Internally legacy parameters are used with new API as part of migration procedure.
      * This fallback can be removed as soon as migration completed */