This is the bug/patch tracker for Rockbox. Click here for more information.
Quick links: Bugs · Patches · Rockbox frontpage
FS#12241 - Voice menus stop working after return to menu from recording screen
Attached to Project:
Rockbox
Opened by Michael Sevakis (MikeS) - Monday, 22 August 2011, 20:35 GMT+2
Last edited by Thomas Martitz (kugel.) - Wednesday, 26 October 2011, 20:32 GMT+2
Opened by Michael Sevakis (MikeS) - Monday, 22 August 2011, 20:35 GMT+2
Last edited by Thomas Martitz (kugel.) - Wednesday, 26 October 2011, 20:32 GMT+2
|
DetailsAs the title suggests, voice menus cease to operate after using the recording screen. A reboot is required to restore voice menu functionality. I checked using r30339 on Gigabeat S and Sansa Fuze v2 thus suspect any SWCODEC target with recording would be affected. I do not at this point know which commit introduced the issue or if further voice issues exist.
|
This task depends upon
Closed by Thomas Martitz (kugel.)
Wednesday, 26 October 2011, 20:32 GMT+2
Reason for closing: Fixed
Additional comments about closing: r30840
Wednesday, 26 October 2011, 20:32 GMT+2
Reason for closing: Fixed
Additional comments about closing: r30840
Occasionally, it seems voice stops working altogether.
Bwahahaha! I had english-us selected as the language instead of english, which was preventing voice working at all. However, the problem reported in this task is still valid.
This seems to possibly affect anything calling talk_disable() to temporarily disable voice that grabs the audio buffer and actually uses it. The same happens with mpegplayer and rockdoom as with recording. Plugins that simply use plugin_get_audio_buffer to stop audio along with talk_disable to disable voice appear not to affect anything.
Is talk_disable even needed any more or was it required merely due to some shared resource? It was (effectively) introduced in r10789, but I don't understand the comment " /* recording_menu gets messed up: so reset talk_menu */" .
It is still needed at this point to keep voice from using stolen buffer space that isn't valid to use.
Ah, I'd assumed the buflib mods had sorted that one out...
Things in that area haven't yet been converted everywhere they could be.
I could reproduce on the sim after adding the pcm_rec_init() call to the sim's main(). Will have a look later, but my suspicion is that voice isn't properly re-initialized after a audio_get_buffer() call.