FS#73690 - Patch qt5-speech to fix order of voices from speech-dispatcher for better default voice.
Attached to Project:
Arch Linux
Opened by Jeremy Whiting (jpwhiting) - Tuesday, 08 February 2022, 19:40 GMT
Last edited by Antonio Rojas (arojas) - Tuesday, 08 February 2022, 21:16 GMT
Opened by Jeremy Whiting (jpwhiting) - Tuesday, 08 February 2022, 19:40 GMT
Last edited by Antonio Rojas (arojas) - Tuesday, 08 February 2022, 21:16 GMT
|
Details
Description:
There's an existing bug in qt5-speech (in upstream code, change submitted for Qt6 here [1]) but since Qt5 has reached end of life and will not be getting any more releases upstream I'm sending this here for inclusion in archlinux directly. May do similarly in debian/other distributions as needed. Here's the problem: When a gui uses QtSpeech library (and on linux uses speech-dispatcher plugin) the order of voices given is reversed by QtSpeech's use of QMultiMap. Since if no voice is explicitly chosen the top voice is selected. This results in defaulting to English (US)+granny if espeak is the only synthesizer set up with speech-dispatcher. rather than English (US). Additional info: * package version(s) * config and/or log files etc. * link to upstream bug report, if any Steps to reproduce: Run kmouth or okular. Both of which in latest versions don't let you select a voice (but will soon). Speak some text or read the current page with okular. Notice the voice sounds like an old woman, not a robot like the default espeak voice. 1: https://codereview.qt-project.org/c/qt/qtspeech/+/394587 |
This task depends upon
Thanks, Closing this one.