Update articles/cognitive-services/Speech-Service/speech-synthesis-markup-voice.md

Co-authored-by: Eric Urban <[email protected]>

Author: sally-baolian

articles/cognitive-services/Speech-Service/speech-synthesis-markup-voice.md

@@ -30,7 +30,7 @@ Usage of the `voice` element's attributes are described in the following table.
 | Attribute | Description | Required or optional |
 | ---------- | ---------- | ---------- |
 | `name`    | The voice used for text-to-speech output. For a complete list of supported prebuilt voices, see [Language support](language-support.md?tabs=tts).| Required|
-| `effect` |A voice-specific effect processor. You can choose a specific value according to the corresponding scenarios. The following values are supported:<br/><ul><li>`eq_car` – Optimize the auditory experience when providing high-fidelity speech in the car scenarios, such as small cars, buses, and other enclosed small/medium vehicles.</li><li>`eq_telecomhp8k` – Optimize the auditory experience in telecom or telephone scenarios. This feature is only designed for narrowband speech (sampling rate = 8kHz). If the sample rate of the output speech is not 8kHz, the auditory quality of the output speech isn't guaranteed even with this attribute. We recommend that you convert the sample rate of the output speech to 8kHz to get a better result with this attribute in telecom scenarios. </li></ul><br/>If the value is missing or invalid, the `effect` attribute will be ignored and the service will use the default neutral speech.|  Optional |
+| `effect` |The audio effect processor that's used to optimize the quality of the synthesized speech output for specific scenarios on devices. <br/><br/>For some scenarios in production environments, the auditory experience may be degraded due to the playback distortion on certain devices. For example, the synthesized speech from a car speaker may sound dull and muffled due to environmental factors such as speaker response, room reverberation, and background noise. The passenger might have to turn up the volume to hear more clearly. To avoid manual operations in such a scenario, the audio effect processor can make the sound clearer by compensating the distortion of playback.<br/><br/>The following values are supported:<br/><ul><li>`eq_car` – Optimize the auditory experience when providing high-fidelity speech in cars, buses, and other enclosed automobiles.</li><li>`eq_telecomhp8k` – Optimize the auditory experience for narrowband speech in telecom or telephone scenarios. We recommend a sampling rate of 8 kHz. If the sample rate isn't 8 kHz, the auditory quality of the output speech won't be optimized.</li></ul><br/>If the value is missing or invalid, this attribute will be ignored and no effect will be applied.|  Optional |
 
 ### Voice examples