THIS IS OUTRAGEOUS. THERE IS A NEW VERSION OF Cepstral Voices WITH AN UPGRADE FEE. HOWEVER, THERE ARE PAGES AND PAGES OF ISSUES ON THE CEPSTRAL WEBSITE THAT SHOW THAT THIS VERSION 5 DOESN'T WORK and can crash your system. THE VERSION I PAID FOR STOPPED WORKING AND THE NEW VERSION BARELY, BARELY WORKS. WHO WANTS THAT??? THIS SHOWS ALL THE WAYS THAT THE NEW VERSION STILL DOESN'T WORK. IF CEPSTRAL CAN'T SPEND THE TIME AND MONEY TO MAKE THEIR PRODUCT WORK WITH THE NEWEST MACOS, THEY SHOULD ABANDON IT AND STOP RIPPING PEOPLE OFF WHO DON';T KNOW WHAT THEY'RE PAYING FOR!!! HERE IS AN ABBREVIATED LIST OF DEFICIENCIES IN THIS SOFTWARE THAT IS POATED ON THE COMPANY'S OWN WEBSITE. AT LEAST THE COMPANY IS BEING UPFRONT BUT THEY STILL WANT TO TAKE YOUR MONEY AND GIVE YOU JUNK IN RETURN.
1. The voices may go silent due to the system's audio properties having been adjusted to incompatible settings.
2. Cepstral voices stop speaking sometimes when used in VoiceOver. They will not speak again through VoiceOver until you log out and log back in. This problem occurs mostly if you use Command+Tab to switch between tasks or if you use Ctrl+Option+D to move the cursor to the dock and cycle through dock items.
3. When reading text copied from Safari, line breaks may occasionally be pronounced as question marks.
4. Word highlighting text positions are wrong when the volume is not set to 1.0.
5. Text position information passed to user callbacks is incorrect with certain settings, causing word highlighting to break.
6. Apple's pronunciation dictionaries are not supported. You can still use cross-platform Cepstral user lexicons, however.
7. Synthesis to phonemes -- the TextToPhonemes() API call -- is not supported.
Error and synchronization callbacks are not supported.
8. When stopping speech, the synthesizer will always halt immediately, instead of waiting until the next word or sentence when requested.
9. Tuned phonetic input -- the [[inpt tune]] embedded command -- is ignored.
In phonemes mode, all prodosic control symbols, with the exception of primary syllable stress, are ignored.
10. Some linebreaks are pronounced as "dot question mark."
11. If you use the Migration Assisant to transfer data and programs from one Mac to another, including Cepstral voices, your Cepstral voices may not function properly on the new computer.
12. If you enable "Announce the Time" under Date & Time preferences and set it to use a Cepstral voice, you cannot change the voice afterwards, even to a different Cepstral voice. Even if you set it to "Use System Voice," it continues to use your selection. Logging out and/or restarting do not solve the problem.
13. "interpret-as" attributes of "" elements are not properly handled.
If the input text contains back-to-back SSML elements, swift only handles the 14. last one in the list. For instance, the following only pauses for three seconds instead of five:
15. RIFF audio files with a sampling rate that differs from the native sampling rate of the voice are not correcly resampled when specified using the SSML element.
16. Applications using Swift TTS exit silently when a sentence that begins with an SSML or tag is fed to a non-US English voice for synthesis.
17. Swift does not currently support the xml:lang attribute of SSML tags that make use of it, such as , , etc.
18. Swift does not reject bad SSML markup. It will complain about blatant syntactic errors, such as missing close tags, but not about semantic errors, such as invalid tags and attributes.
19. The SSML and tags may not have the proper effect. This is supposed to be read as two sentences, but we read it as one:
20. Swift does not currently support the SSML tag for adding additional lexicons.
21. The SSML tag:attribute pair should select a neuter-gendered voice, but it currently does nothing.
22. The "age" attribute to the SSML tag is looking for an exact match, when it probably ought to be looking for the closest match. For instance, will only select a ten-year-old voice, or fall back to the default voice if one is not found.
23. Cannot provide multiple voice selections in the SSML tag. The tag should request David, and use William as a fallback. Instead, Swift interprets as a request for a voice named "David William."
24. The default SSML voice may not be the default voice for that gender. To illustrate, Swift may switch voices while executing the following SSML code, when it should not.
25. Changing voice in SSML resets prosody and pitch.
26. The duration attribute of the SSML tag is not supported.
27. Prosody as specified in the SSML prosody element is not applied to output generated using the SSML phoneme element.
28. Pitch contours passed in via the SSML tag with values not falling between 0% and 100% are supposed to be ignored. Swift attempts to synthesize with these bogus values.
29. SAMPA stress markers in SSML tag not supported.
30. The SSML tag causes a crash in Canadian French, Americas Spanish, and Italian voices.
40. Text position synthesis event information is incorrect when the SSML element is encountered.