[DEV-WIP] MIC testing (To enable GoogleNow function in JB)

Search This thread
By:
Advanced…
Hashcode

Hashcode

Senior Recognized Developer
Sep 3, 2011
3,426
23,732
NOTE: This requires some knowledge of the ALSA subsystem. If you're not comfortable doing this, PLEASE DONT. You can in theory damage the audio hardware with bad settings.

With the migration to Jelly Bean, we can see how powerful having a working MIC would be.

On my latest AOSP Jelly Bean ROM (7/13), I have 3 new binaries so that anyone who is comfortable with the audio system can jump in and help:
tinymix
tinycap
tinyplay

Also, you can install "AlsaMixer" from the Market which gives you a bit of control over the audio system's low level devices.

I feel like if we can get a few devs / knowledgeable users looking at these settings, that we could identify where in our audio driver the mixer for the MIC is getting off track.

THE GOAL:
Get a small list of mixer changes (tinymix / AlsaMixer) which enable the MIC for use in the OS.
Then I can replicate those in the audio driver.


GAME ON.
 
Last edited:
  • Like
Reactions: donnelpa, quan_1986, Guitarman and 27 others
times_infinity

times_infinity

Senior Member
Oct 26, 2010
1,534
1,613
Samsung Galaxy S9
Samsung Galaxy S22 Ultra
Hashcode said:
NOTE: This requires some knowledge of the ALSA subsystem. If you're not comfortable doing this, PLEASE DONT. You can in theory damage the audio hardware with bad settings.

With the migration to Jelly Bean, we can see how powerful having a working MIC would be.

On my latest AOSP Jelly Bear ROM (7/13), I have 3 new binaries so that anyone who is comfortable with the audio system can jump in and help:
tinymix
tinycap
tinyplay

Also, you can install "AlsaMixer" from the Market which gives you a bit of control over the audio system's low level devices.

I feel like if we can get a few devs / knowledgeable users looking at these settings, that we could identify where in our audio driver the mixer for the MIC is getting off track.

THE GOAL:
Get a small list of mixer changes (tinymix / AlsaMixer) which enable the MIC for use in the OS.
Then I can replicate those in the audio driver.


GAME ON.
Click to expand...
Click to collapse

Perfect person to probably ask for help would be supercurio, although his specialty is wolfson codecs
 
webdroidmt

webdroidmt

Senior Member
Dec 20, 2011
1,674
1,156
NJ
Hashcode said:
NOTE: This requires some knowledge of the ALSA subsystem. If you're not comfortable doing this, PLEASE DONT. You can in theory damage the audio hardware with bad settings.

With the migration to Jelly Bean, we can see how powerful having a working MIC would be.

On my latest AOSP Jelly Bear ROM (7/13), I have 3 new binaries so that anyone who is comfortable with the audio system can jump in and help:
tinymix
tinycap
tinyplay

Also, you can install "AlsaMixer" from the Market which gives you a bit of control over the audio system's low level devices.

I feel like if we can get a few devs / knowledgeable users looking at these settings, that we could identify where in our audio driver the mixer for the MIC is getting off track.

THE GOAL:
Get a small list of mixer changes (tinymix / AlsaMixer) which enable the MIC for use in the OS.
Then I can replicate those in the audio driver.


GAME ON.
Click to expand...
Click to collapse

This would be great! Btw, is your rom's codename Jelly Bear? :D

Mike T
 
R

roothorick

Senior Member
May 2, 2012
348
98
I don't think the Google Now crash is related to the microphone....


Here's the logcat I'm getting:
Code: 
W/dalvikvm( 3429): threadid=1: thread exiting with uncaught exception (group=0x41199300)
E/AndroidRuntime( 3429): FATAL EXCEPTION: main
E/AndroidRuntime( 3429): java.lang.IllegalArgumentException: requested provider network doesn't exisit
E/AndroidRuntime( 3429): 	at android.os.Parcel.readException(Parcel.java:1429)
E/AndroidRuntime( 3429): 	at android.os.Parcel.readException(Parcel.java:1379)
E/AndroidRuntime( 3429): 	at android.location.ILocationManager$Stub$Proxy.requestLocationUpdates(ILocationManager.java:646)
E/AndroidRuntime( 3429): 	at android.location.LocationManager._requestLocationUpdates(LocationManager.java:660)
E/AndroidRuntime( 3429): 	at android.location.LocationManager.requestLocationUpdates(LocationManager.java:482)
E/AndroidRuntime( 3429): 	at com.google.android.apps.sidekick.inject.SystemLocationManagerInjectable.requestLocationUpdates(SystemLocationManagerInjectable.java:33)
E/AndroidRuntime( 3429): 	at com.google.android.apps.sidekick.LocationOracleImpl.registerForAndroidNlpFix(LocationOracleImpl.java:495)
E/AndroidRuntime( 3429): 	at com.google.android.apps.sidekick.LocationOracleImpl.start(LocationOracleImpl.java:139)
E/AndroidRuntime( 3429): 	at com.google.android.velvet.VelvetApplication.startTgServices(VelvetApplication.java:175)
E/AndroidRuntime( 3429): 	at com.google.android.velvet.tg.FirstRunActivity.completeOptIn(FirstRunActivity.java:350)
E/AndroidRuntime( 3429): 	at com.google.android.velvet.tg.FirstRunActivity$OptInHander.handleMessage(FirstRunActivity.java:421)
E/AndroidRuntime( 3429): 	at android.os.Handler.dispatchMessage(Handler.java:99)
E/AndroidRuntime( 3429): 	at android.os.Looper.loop(Looper.java:137)
E/AndroidRuntime( 3429): 	at android.app.ActivityThread.main(ActivityThread.java:4745)
E/AndroidRuntime( 3429): 	at java.lang.reflect.Method.invokeNative(Native Method)
E/AndroidRuntime( 3429): 	at java.lang.reflect.Method.invoke(Method.java:511)
E/AndroidRuntime( 3429): 	at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:786)
E/AndroidRuntime( 3429): 	at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:553)
E/AndroidRuntime( 3429): 	at dalvik.system.NativeStart.main(Native Method)
W/ActivityManager(  257):   Force finishing activity com.google.android.googlequicksearchbox/com.google.android.velvet.tg.FirstRunActivity

That doesn't look microphone-related. I think our smoking gun is Wi-Fi location. Unfortunately, disabling location services didn't help :/ Hash, when you get the 2nd part of Wi-Fi location working, try Now and see what it does.
 
  • Like
Reactions: tobiascuypers, Jaimemex and MagicXB
Jaimemex

Jaimemex

Senior Member
Jan 11, 2010
407
55
roothorick said:
I don't think the Google Now crash is related to the microphone....


Here's the logcat I'm getting:
Code: 
W/dalvikvm( 3429): threadid=1: thread exiting with uncaught exception (group=0x41199300)
E/AndroidRuntime( 3429): FATAL EXCEPTION: main
E/AndroidRuntime( 3429): java.lang.IllegalArgumentException: requested provider network doesn't exisit
E/AndroidRuntime( 3429): at android.os.Parcel.readException(Parcel.java:1429)
E/AndroidRuntime( 3429): at android.os.Parcel.readException(Parcel.java:1379)
E/AndroidRuntime( 3429): at android.location.ILocationManager$Stub$Proxy.requestLocationUpdates(ILocationManager.java:646)
E/AndroidRuntime( 3429): at android.location.LocationManager._requestLocationUpdates(LocationManager.java:660)
E/AndroidRuntime( 3429): at android.location.LocationManager.requestLocationUpdates(LocationManager.java:482)
E/AndroidRuntime( 3429): at com.google.android.apps.sidekick.inject.SystemLocationManagerInjectable.requestLocationUpdates(SystemLocationManagerInjectable.java:33)
E/AndroidRuntime( 3429): at com.google.android.apps.sidekick.LocationOracleImpl.registerForAndroidNlpFix(LocationOracleImpl.java:495)
E/AndroidRuntime( 3429): at com.google.android.apps.sidekick.LocationOracleImpl.start(LocationOracleImpl.java:139)
E/AndroidRuntime( 3429): at com.google.android.velvet.VelvetApplication.startTgServices(VelvetApplication.java:175)
E/AndroidRuntime( 3429): at com.google.android.velvet.tg.FirstRunActivity.completeOptIn(FirstRunActivity.java:350)
E/AndroidRuntime( 3429): at com.google.android.velvet.tg.FirstRunActivity$OptInHander.handleMessage(FirstRunActivity.java:421)
E/AndroidRuntime( 3429): at android.os.Handler.dispatchMessage(Handler.java:99)
E/AndroidRuntime( 3429): at android.os.Looper.loop(Looper.java:137)
E/AndroidRuntime( 3429): at android.app.ActivityThread.main(ActivityThread.java:4745)
E/AndroidRuntime( 3429): at java.lang.reflect.Method.invokeNative(Native Method)
E/AndroidRuntime( 3429): at java.lang.reflect.Method.invoke(Method.java:511)
E/AndroidRuntime( 3429): at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:786)
E/AndroidRuntime( 3429): at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:553)
E/AndroidRuntime( 3429): at dalvik.system.NativeStart.main(Native Method)
W/ActivityManager(  257):   Force finishing activity com.google.android.googlequicksearchbox/com.google.android.velvet.tg.FirstRunActivity

That doesn't look microphone-related. I think our smoking gun is Wi-Fi location. Unfortunately, disabling location services didn't help :/ Hash, when you get the 2nd part of Wi-Fi location working, try Now and see what it does.
Click to expand...
Click to collapse

Even if the mic isn't the issue I would still like to see this implemented this would let many more apps function like shazam, voice typing and of course voice search...

Sent from my Amazon Kindle Fire using xda premium
 
Hashcode

Hashcode

Senior Recognized Developer
Sep 3, 2011
3,426
23,732
roothorick said:
I don't think the Google Now crash is related to the microphone....


Here's the logcat I'm getting:
Code: 
{removed}

That doesn't look microphone-related. I think our smoking gun is Wi-Fi location. Unfortunately, disabling location services didn't help :/ Hash, when you get the 2nd part of Wi-Fi location working, try Now and see what it does.
Click to expand...
Click to collapse

Ah, I was seeing an "AudioRecord" error talking about trying to use a non-existant device. This error above I can work on.
 
  • Like
Reactions: donnelpa, vistazifta, chrisliphart and 3 others
dadeo1111

dadeo1111

Senior Member
Jul 6, 2011
157
49
http://en.wikipedia.org/wiki/Springfield
Hashcode said:
NOTE: This requires some knowledge of the ALSA subsystem. If you're not comfortable doing this, PLEASE DONT. You can in theory damage the audio hardware with bad settings.

With the migration to Jelly Bean, we can see how powerful having a working MIC would be.

On my latest AOSP Jelly Bean ROM (7/13), I have 3 new binaries so that anyone who is comfortable with the audio system can jump in and help:
tinymix
tinycap
tinyplay

Also, you can install "AlsaMixer" from the Market which gives you a bit of control over the audio system's low level devices.

I feel like if we can get a few devs / knowledgeable users looking at these settings, that we could identify where in our audio driver the mixer for the MIC is getting off track.

THE GOAL:
Get a small list of mixer changes (tinymix / AlsaMixer) which enable the MIC for use in the OS.
Then I can replicate those in the audio driver.


GAME ON.
Click to expand...
Click to collapse

No ALSA experience other than ticking a box or two on my Puppy Linux install - sure wish I did.

This needs to be nailed down for Jelly Bean to work well.

bump.jpg
 
R

roothorick

Senior Member
May 2, 2012
348
98
Hashcode said:
Ah, I was seeing an "AudioRecord" error talking about trying to use a non-existant device. This error above I can work on.
Click to expand...
Click to collapse

This is purely speculation, but did the other stacktrace come from the 7/11 build? Maybe your half-fix for Wi-Fi location actually broke things, and we actually have TWO blockers here. We won't know for sure until you get location working all the way though.
 
  • Like
Reactions: Jaimemex
chupchae

chupchae

Senior Member
Jul 24, 2006
82
7
Skopje
chupchae.tumblr.com
Alsa mixer keeps installing

I had no luck trying to experiment with alsa mixer sound variables - I installed it from store successfully, but anytime I leave alsa mixer in a way to check is external mic working with some recording app, when I come back to alsa mixer it demands to be again installed. Someone went further with alsa mixer on Kindle JB?
 
S

stylon

Member
Oct 17, 2011
48
60
Aachen, Germany
hack-a-fire.blogspot.com
Hashcode said:
On my latest AOSP Jelly Bean ROM (7/13), I have 3 new binaries so that anyone who is comfortable with the audio system can jump in and help:
tinymix
tinycap
tinyplay
Click to expand...
Click to collapse

Hi Hash,

I've no idea wether anybody else tried, but I just did on your CM10.1 ROM (actually no nightly, I've build one myself as I wanted to learn more about the build process, but it's very close to the latest nightly). And guess what... 2 minutes into trying... it WORKS!!!!! :good::good::good:

Here's what I did:
tinycap xxx.wav -d 3 -c 1 -r 16000 -b 16

I used some regular white Apple iPhone Headset I had lying around (those with the 4-ring plug and the mic in the vol +/- housing).

Except for trying the other capture devices, I promise... I did nothing else, not even played around with tinymix. -d 3 was important... 0 doesn't have capture and the other capture devices didn't even like these simple settings. Now, let's get the audio routed to Google Voice Search ;).

Best,
STYLON

P.S.: please check out my backlight fix... :D I fear that what we've there now drains the batteries as even though it seems to be off (screen really completely dark), it's not off very often for me.
 
  • Like
Reactions: Ichigo, Guitarman and BrooklynNY
BrooklynNY

BrooklynNY

Senior Member
Nov 16, 2012
374
85
stylon said:
Hi Hash, I've no idea wether anybody else tried, but I just did on your CM10.1 ROM (actually no nightly, I've build one myself as I wanted to learn more about the build process, but it's very close to the latest nightly). And guess what... 2 minutes into trying... it WORKS!!!!! :good::good::good:
Here's what I did: tinycap xxx.wav -d 3 -c 1 -r 16000 -b 16......
Click to expand...
Click to collapse
Can it be done with this ROM?
 
Last edited:
A

AndroidSlave

Guest
stylon said:
Hi Hash,

I've no idea wether anybody else tried, but I just did on your CM10.1 ROM (actually no nightly, I've build one myself as I wanted to learn more about the build process, but it's very close to the latest nightly). And guess what... 2 minutes into trying... it WORKS!!!!! :good::good::good:

Here's what I did:
tinycap xxx.wav -d 3 -c 1 -r 16000 -b 16

I used some regular white Apple iPhone Headset I had lying around (those with the 4-ring plug and the mic in the vol +/- housing).

Except for trying the other capture devices, I promise... I did nothing else, not even played around with tinymix. -d 3 was important... 0 doesn't have capture and the other capture devices didn't even like these simple settings. Now, let's get the audio routed to Google Voice Search ;).

Best,
STYLON

P.S.: please check out my backlight fix... :D I fear that what we've there now drains the batteries as even though it seems to be off (screen really completely dark), it's not off very often for me.
Click to expand...
Click to collapse

I'm not sure hashcode follows this thread anymore...you may want to PM him or try him on another thread

Sent from my GT-N7000 using Tapatalk 2
 
You must log in or register to reply here.

Similar threads

twa_priv
[ROM] [JB 4.2.2] CM10.1/SGT7 for the Kindle Fire [20130724]
Replies
2K
Views
630K
t3chwizard
t3chwizard
bl1nkk
[ROM][24 Feb] IceCreamSandwich Development/Discussion Thread
Replies
3K
Views
1M
H
heyjoe66
Hashcode
  • Locked
Letting this die. New dev and user threads created.
Replies
346
Views
159K
B
boomn
powerpoint45
[ROM] [CM10] [JB] ▶Alien Droid◀(Complete GUI customization) [12.15.12][DISCONTINUED]
Replies
584
Views
157K
powerpoint45
powerpoint45
NRGZ28
[ROM][ICS] |July 31| Energy™ -.¸¸.·´¯ ICS CM9 3.0 Kernel / Lightly Themed ¯´·.¸¸. -
Replies
1K
Views
500K
S
samreh

Top Liked Posts

24 Hours All time
  • There are no posts matching your filters.
  • 30
    Hashcode
    Hashcode
    NOTE: This requires some knowledge of the ALSA subsystem. If you're not comfortable doing this, PLEASE DONT. You can in theory damage the audio hardware with bad settings.

    With the migration to Jelly Bean, we can see how powerful having a working MIC would be.

    On my latest AOSP Jelly Bean ROM (7/13), I have 3 new binaries so that anyone who is comfortable with the audio system can jump in and help:
    tinymix
    tinycap
    tinyplay

    Also, you can install "AlsaMixer" from the Market which gives you a bit of control over the audio system's low level devices.

    I feel like if we can get a few devs / knowledgeable users looking at these settings, that we could identify where in our audio driver the mixer for the MIC is getting off track.

    THE GOAL:
    Get a small list of mixer changes (tinymix / AlsaMixer) which enable the MIC for use in the OS.
    Then I can replicate those in the audio driver.


    GAME ON.
    View
    11
    Hashcode
    Hashcode
    It's the part of the Android operating system that interacts directly with the kernel software driver. This piece is changed out (potentially) for each device where the rest of Android's audio system is the same.

    The audio changes aren't just for the MIC alone. I'm working on finishing the kernel 3.4 upgrade and the source in the kernel for the audio codec chip is a complete re-write.
    View
    8
    Hashcode
    Hashcode
    Hi Stylon,

    Got the PM. I'm buried at the moment developing a 3.4 kernel for the RAZR-HD. However, I did plan to revisit backlight / light sensor changes and the audio issues once I free up. (As well as the 3.4 kernel for the KFire's -- tho it may be a 3.8 kernel by the look of things.)

    Great work on testing the audio input w/ tinycap. :)
    View
    7
    S
    sbradymobile
    stylon said:
    Hi Hash,

    I've no idea wether anybody else tried, but I just did on your CM10.1 ROM (actually no nightly, I've build one myself as I wanted to learn more about the build process, but it's very close to the latest nightly). And guess what... 2 minutes into trying... it WORKS!!!!! :good::good::good:

    Here's what I did:
    tinycap xxx.wav -d 3 -c 1 -r 16000 -b 16

    I used some regular white Apple iPhone Headset I had lying around (those with the 4-ring plug and the mic in the vol +/- housing).

    Except for trying the other capture devices, I promise... I did nothing else, not even played around with tinymix. -d 3 was important... 0 doesn't have capture and the other capture devices didn't even like these simple settings. Now, let's get the audio routed to Google Voice Search ;).

    Best,
    STYLON

    P.S.: please check out my backlight fix... :D I fear that what we've there now drains the batteries as even though it seems to be off (screen really completely dark), it's not off very often for me.
    Click to expand...
    Click to collapse

    Thanks, with the changing the device from 2 to 3 (see below) it looks like Google Now will work. I haven't tried any other apps yet, but you can give this a try in device/amazon/otter-common for testing.

    diff --git a/audio/audio_hw.c b/audio/audio_hw.c
    index 22ebe47..7cc597d 100644
    --- a/audio/audio_hw.c
    +++ b/audio/audio_hw.c
    @@ -2245,7 +2245,7 @@ static int start_input_stream(struct omap_stream_in *in)
    {
    int ret = 0;
    unsigned int card = CARD_OMAP_DEFAULT;
    - unsigned int device = PORT_MM2_UL;
    + unsigned int device = PORT_MM_LEGACY;
    struct omap_audio_device *adev = in->dev;
    unsigned int vx_rec_ul_on = (in->source == AUDIO_SOURCE_VOICE_UPLINK) ||
    (in->source == AUDIO_SOURCE_VOICE_CALL);

    and use the audio_policy.conf from the audio directory.

    Attached is a zip for those interested in testing.
    View
    6
    Hashcode
    Hashcode
    roothorick said:
    I don't think the Google Now crash is related to the microphone....


    Here's the logcat I'm getting:
    Code: 
    {removed}

    That doesn't look microphone-related. I think our smoking gun is Wi-Fi location. Unfortunately, disabling location services didn't help :/ Hash, when you get the 2nd part of Wi-Fi location working, try Now and see what it does.
    Click to expand...
    Click to collapse

    Ah, I was seeing an "AudioRecord" error talking about trying to use a non-existant device. This error above I can work on.
    View

New posts