You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
| first byte latency | Indicates the time delay between the start of the synthesis task and receipt of the first chunk of audio data. | SpeechServiceResponse_SynthesisFirstByteLatencyMs |
28
-
| finish latency | Indicates the time delay between the start of the synthesis task and the receipt of the whole synthesized audio data. | SpeechServiceResponse_SynthesisFinishLatencyMs |
27
+
|`first byte client latency`| Indicates the time delay between the synthesis starts and the first audio chunk is received on the client including network latency.|`SpeechServiceResponse_SynthesisFirstByteLatencyMs`|
28
+
|`finish client latency`| Indicates the time delay between the synthesis starts and the whole synthesized audio is received on the client including network latency. |`SpeechServiceResponse_SynthesisFinishLatencyMs`|
29
+
|`network latency`| The network latency between the client and Azure TTS service. |`SpeechServiceResponse_SynthesisNetworkLatencyMs`|
30
+
|`first byte service latency`| Indicates the time delay between Azure TTS service received synthesis request and the first audio chunk is returned. |`SpeechServiceResponse_SynthesisServiceLatencyMs`|
29
31
30
32
The Speech SDK puts the latency durations in the Properties collection of [`SpeechSynthesisResult`](/dotnet/api/microsoft.cognitiveservices.speech.speechsynthesisresult). The following sample code shows these values.
|`first byte latency`| Indicates the time delay between the synthesis starts and the first audio chunk is received. |`SpeechServiceResponse_SynthesisFirstByteLatencyMs`|
47
-
|`finish latency`| Indicates the time delay between the synthesis starts and the whole synthesized audio is received. |`SpeechServiceResponse_SynthesisFinishLatencyMs`|
50
+
|`first byte client latency`| Indicates the time delay between the synthesis starts and the first audio chunk is received on the client including network latency.|`SpeechServiceResponse_SynthesisFirstByteLatencyMs`|
51
+
|`finish client latency`| Indicates the time delay between the synthesis starts and the whole synthesized audio is received on the client including network latency. |`SpeechServiceResponse_SynthesisFinishLatencyMs`|
52
+
|`network latency`| The network latency between the client and Azure TTS service. |`SpeechServiceResponse_SynthesisNetworkLatencyMs`|
53
+
|`first byte service latency`| Indicates the time delay between Azure TTS service received synthesis request and the first audio chunk is returned. |`SpeechServiceResponse_SynthesisServiceLatencyMs`|
48
54
49
55
The Speech SDK measured the latencies and puts them in the property bag of [`SpeechSynthesisResult`](/cpp/cognitive-services/speech/speechsynthesisresult). Refer following codes to get them.
50
56
51
57
```cpp
52
58
auto result = synthesizer->SpeakTextAsync(text).get();
53
59
auto firstByteLatency = std::stoi(result->Properties.GetProperty(PropertyId::SpeechServiceResponse_SynthesisFirstByteLatencyMs));
54
60
auto finishedLatency = std::stoi(result->Properties.GetProperty(PropertyId::SpeechServiceResponse_SynthesisFinishLatencyMs));
61
+
auto firstByteLatency = std::stoi(result->Properties.GetProperty(PropertyId::SpeechServiceResponse_SynthesisNetworkLatencyMs));
62
+
auto firstByteLatency = std::stoi(result->Properties.GetProperty(PropertyId::SpeechServiceResponse_SynthesisServiceLatencyMs));
55
63
// you can also get the result id, and send to us when you need help for diagnosis
56
64
auto resultId = result->ResultId;
57
65
```
@@ -62,15 +70,19 @@ auto resultId = result->ResultId;
| `first byte latency` | Indicates the time delay between the synthesis starts and the first audio chunk is received. | `SpeechServiceResponse_SynthesisFirstByteLatencyMs` |
66
-
| `finish latency` | Indicates the time delay between the synthesis starts and the whole synthesized audio is received. | `SpeechServiceResponse_SynthesisFinishLatencyMs` |
73
+
| `first byte client latency` | Indicates the time delay between the synthesis starts and the first audio chunk is received on the client including network latency.| `SpeechServiceResponse_SynthesisFirstByteLatencyMs` |
74
+
| `finish client latency` | Indicates the time delay between the synthesis starts and the whole synthesized audio is received on the client including network latency. | `SpeechServiceResponse_SynthesisFinishLatencyMs` |
75
+
| `network latency` | The network latency between the client and Azure TTS service. | `SpeechServiceResponse_SynthesisNetworkLatencyMs` |
76
+
| `first byte service latency` | Indicates the time delay between Azure TTS service received synthesis request and the first audio chunk is returned. | `SpeechServiceResponse_SynthesisServiceLatencyMs` |
67
77
68
78
The Speech SDK measured the latencies and puts them in the property bag of [`SpeechSynthesisResult`](/java/api/com.microsoft.cognitiveservices.speech.speechsynthesisresult). Refer following codes to get them.
69
79
70
80
```java
71
81
SpeechSynthesisResult result = synthesizer.SpeakTextAsync(text).get();
|`first byte latency`| Indicates the time delay between the synthesis starts and the first audio chunk is received. |`SpeechServiceResponse_SynthesisFirstByteLatencyMs`|
86
-
|`finish latency`| Indicates the time delay between the synthesis starts and the whole synthesized audio is received. |`SpeechServiceResponse_SynthesisFinishLatencyMs`|
97
+
|`first byte client latency`| Indicates the time delay between the synthesis starts and the first audio chunk is received on the client including network latency.|`SpeechServiceResponse_SynthesisFirstByteLatencyMs`|
98
+
|`finish client latency`| Indicates the time delay between the synthesis starts and the whole synthesized audio is received on the client including network latency. |`SpeechServiceResponse_SynthesisFinishLatencyMs`|
99
+
|`network latency`| The network latency between the client and Azure TTS service. |`SpeechServiceResponse_SynthesisNetworkLatencyMs`|
100
+
|`first byte service latency`| Indicates the time delay between Azure TTS service received synthesis request and the first audio chunk is returned. |`SpeechServiceResponse_SynthesisServiceLatencyMs`|
87
101
88
102
The Speech SDK measured the latencies and puts them in the property bag of [`SpeechSynthesisResult`](/python/api/azure-cognitiveservices-speech/azure.cognitiveservices.speech.speechsynthesisresult). Refer following codes to get them.
|`first byte latency`| Indicates the time delay between the synthesis starts and the first audio chunk is received. |`SPXSpeechServiceResponseSynthesisFirstByteLatencyMs`|
105
-
|`finish latency`| Indicates the time delay between the synthesis starts and the whole synthesized audio is received. |`SPXSpeechServiceResponseSynthesisFinishLatencyMs`|
120
+
|`first byte client latency`| Indicates the time delay between the synthesis starts and the first audio chunk is received on the client including network latency. |`SPXSpeechServiceResponseSynthesisFirstByteLatencyMs`|
121
+
|`finish client latency`| Indicates the time delay between the synthesis starts and the whole synthesized audio is received on the client including network latency. |`SPXSpeechServiceResponseSynthesisFinishLatencyMs`|
122
+
|`network latency`| The network latency between the client and Azure TTS service. |`SPXSpeechServiceResponseSynthesisNetworkLatencyMs`|
123
+
|`first byte service latency`| Indicates the time delay between Azure TTS service received synthesis request and the first audio chunk is returned. |`SPXSpeechServiceResponseSynthesisServiceLatencyMs`|
106
124
107
125
The Speech SDK measured the latencies and puts them in the property bag of [`SPXSpeechSynthesisResult`](/objectivec/cognitive-services/speech/spxspeechsynthesisresult). Refer following codes to get them.
0 commit comments