TTS playback garbled

The information in this article applies to:

• SpeechKit 3

Symptoms

When synthesizing text, the audio playback is garbled and cannot be understood.

Cause

The reason for the garbled playback is that the speed is too high. With SAPI5 voices, the speed is a setting actually maps to the ISpVoice Rate that is defined as the value specifying the speaking rate of the voice. Supported values range from -10 to 10 and values outside this range may be truncated.

If you are switching voices, including SAPI4 voices, the old voice properties were being applied to the new voice properties by default. This can cause an inappropriate speed or rate unless you change it by setting the property value.

Status

Fixed in SpeechKit 3 SR2. The application framework setting will be used if previously defined for the specific TTS engine otherwise the TTS engine default voice settings are used.

Component Formats Impacted

All formats.