1ALSABAT(1) General Commands Manual ALSABAT(1)
2
3
4
6 alsabat - command-line sound tester for ALSA sound card driver
7
8
10 alsabat [flags]
11
12
14 ALSABAT(ALSA Basic Audio Tester) is a simple command-line utility
15 intended to help automate audio driver and sound server testing with
16 little human interaction. ALSABAT can be used to test audio quality,
17 stress test features and test audio before and after PM state changes.
18
19 ALSABAT's design is relatively simple. ALSABAT plays an audio stream
20 and captures the same stream in either a digital or analog loop back.
21 It then compares the captured stream using a FFT to the original to
22 determine if the test case passes or fails.
23
24 ALSABAT can either run wholly on the target machine being tested
25 (standalone mode) or can run as a client/server mode where by alsabat
26 client runs on the target and runs as a server on a separate tester
27 machine. The client/server mode still requires some manual interaction
28 for synchronization, but this is actively being developed for future
29 releases.
30
31 The hardware testing configuration may require the use of an analog
32 cable connecting target to tester machines or a cable to create an ana‐
33 log loopback if no loopback mode is available on the sound hardware
34 that is being tested. An analog loopback cable can be used to connect
35 the "line in" to "line out" jacks to create a loopback. If only head‐
36 phone and mic jacks (or combo jack) are available then the following
37 simple circuit can be used to create an analog loopback :-
38
39 https://source.android.com/devices/audio/loopback.html
40
41 If tinyalsa is installed in system, user can choose tinyalsa as backend
42 lib of alsabat, with configure option "--enable-alsabat-backend-tiny".
43
44
46 -h, --help
47 Help: show syntax.
48
49 -D Select sound card to be tested by name.
50
51 -P Select the playback PCM device.
52
53 -C Select the capture PCM device.
54
55 -f Sample format
56 Recognized sample formats are: U8 S16_LE S24_3LE S32_LE
57 Some of these may not be available on selected hardware
58 The available format shortcuts are:
59 -f cd (16 bit little endian, 44100, stereo) [-f S16_LE -c2 -r44100]
60 -f dat (16 bit little endian, 48000, stereo) [-f S16_LE -c2 -r48000]
61 If no format is given S16_LE is used.
62
63 -c The number of channels. The default is one channel. Valid val‐
64 ues at the moment are 1 or 2.
65
66 -r Sampling rate in Hertz. The default rate is 44100 Hertz. Valid
67 values depends on hardware support.
68
69 -n Duration of generated signal. The value could be either of the
70 two forms:
71 1. Decimal integer, means number of frames;
72 2. Floating point with suffix 's', means number of seconds.
73 The default is 2 seconds.
74
75 -k Sigma k value for analysis.
76 The analysis function reads data from WAV file, run FFT against
77 the data to get magnitude of frequency vectors, and then calcu‐
78 lates the average value and standard deviation of frequency vec‐
79 tors. After that, we define a threshold:
80 threshold = k * standard_deviation + mean_value
81 Frequencies with amplitude larger than threshold will be recog‐
82 nized as a peak, and the frequency with largest peak value will
83 be recognized as a detected frequency.
84 ALSABAT then compares the detected frequency to target fre‐
85 quency, to decide if the detecting passes or fails.
86 The default value is 3.0.
87
88 -F Target frequency for signal generation and analysis, in Hertz.
89 The default is 997.0 Hertz. Valid range is (DC_THRESHOLD, 40% *
90 Sampling rate).
91
92 -p Total number of periods to play or capture.
93
94 --log=#
95 Write stderr and stdout output to this log file.
96
97 --file=#
98 Input WAV file for playback.
99
100 --saveplay=#
101 Target WAV file to save capture test content.
102
103 --local
104 Internal loopback mode. Playback, capture and analysis internal
105 to ALSABAT only. This is intended for developers to test new
106 ALSABAT features as no audio is routed outside of ALSABAT.
107
108 --standalone
109 Add support for standalone mode where ALSABAT will run on a dif‐
110 ferent machine to the one being tested. In standalone mode, the
111 sound data can be generated, playback and captured just like in
112 normal mode, but will not be analyzed. The ALSABAT being built
113 without libfftw3 support is always in standalone mode. The
114 ALSABAT in normal mode can also bypass data analysis using
115 option "--standalone".
116
117 --roundtriplatency
118 Round trip latency test. Audio latency is the time delay as an
119 audio signal passes through a system. There are many kinds of
120 audio latency metrics. One useful metric is the round trip
121 latency, which is the sum of output latency and input latency.
122
123 --snr-db=#
124 Noise detection threshold in SNR (dB). 26dB indicates 5% noise
125 in amplitude. ALSABAT will return error if signal SNR is
126 smaller than the threshold.
127
128 --snr-pc=#
129 Noise detection threshold in percentage of noise amplitude (%).
130 ALSABAT will return error if the noise amplitude is larger than
131 the threshold.
132
133
135 alsabat -P plughw:0,0 -C plughw:0,0 -c 2 -f S32_LE -F 250
136 Generate and play a sine wave of 250 Hertz with 2 channel and
137 S32_LE format, and then capture and analyze.
138
139
140 alsabat -P plughw:0,0 -C plughw:0,0 --file 500Hz.wav
141 Play the RIFF WAV file "500Hz.wav" which contains 500 Hertz
142 waveform LPCM data, and then capture and analyze.
143
144
146 On success, returns 0.
147 If no peak be detected, returns -1001;
148 If only DC be detected, returns -1002;
149 If peak frequency does not match with the target frequency, returns
150 -1003.
151
152
154 aplay(1)
155
156
158 Currently only support RIFF WAV format with PCM data. Please report any
159 bugs to the alsa-devel mailing list.
160
161
163 alsabat is by Liam Girdwood <liam.r.girdwood@linux.intel.com>, Bernard
164 Gautier <bernard.gautier@intel.com> and Han Lu <han.lu@intel.com>.
165 This document is by Liam Girdwood <liam.r.girdwood@linux.intel.com> and
166 Han Lu <han.lu@intel.com>.
167
168
169
170 20th October 2015 ALSABAT(1)