[kohira]

Hi,

I executed a modified my_sep_rec_offline_2.n NOT by harkmw.exe but on HARK Designer 3.0.0, and got another error shown below.

Should I add any node?

Thank you.
Kohira

—————————-
Stopped
Start HARK middleware:
network: my_sep_rec_offline_2.n
execute: my_sep_rec_offline_2.n
process: local
toolbox: [‘C:/Program Files/HARK/lib/hark/toolbox’]
args: []

reading A matrix

72 srcs, 8 microphones, 512 points

Traceback (most recent call last):
File “C:\ProgramData\Anaconda3\Scripts\batchflow-script.py”, line 11, in <module>
load_entry_point(‘harkmw==3.0.0’, ‘console_scripts’, ‘batchflow’)()
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\main.py”, line 102, in main
app.run(process_name)
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\process.py”, line 71, in run
network.run(self)
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\network.py”, line 46, in run
self.proceed(0, process)
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\network.py”, line 103, in proceed
count = self.proceed_frame(count, process)
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\network.py”, line 134, in proceed_frame
minimum_count = min(minimum_count, node.push_data(None, count, process))
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\node.py”, line 129, in push_data
minimum_target_count = min(minimum_target_count, output_proxy.push_data(target_count, process))
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\proxy.py”, line 180, in push_data
count = self.output_node.push_data(self, count, process)
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\node.py”, line 129, in push_data
minimum_target_count = min(minimum_target_count, output_proxy.push_data(target_count, process))
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\proxy.py”, line 178, in push_data
self.output_node.network.push_data(input, count, process)
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\network.py”, line 225, in push_data
return self.proceed(count, process)
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\logic\network.py”, line 78, in proceed
condition_satisfied = node.get_output(“OUTPUT”, internal_count)
File “C:\ProgramData\Anaconda3\lib\site-packages\harkmw\types\native_node.py”, line 126, in get_output
return self.native_node.get_output(name, count)
RuntimeError: HARK-Middleware: HARK exception thrown at harkmw/src/node.cc:215 (Node_get_output): Cast error: Trying to cast ObjectRef (N4HARK7FILEPTRE) into N4HARK7IStreamE
C:\msys64\home\NAE2\work\src\hark-base\hark-base\src\BufferedNode.cc line 98: Node AudioStreamFromWave (type 19AudioStreamFromWave) Exception caught in BufferedNode::getOutput
C:\msys64\home\NAE2\work\src\hark-base\hark-base\src\BufferedNode.cc line 98: Node ProxyNode (type N6HARKMW9ProxyNodeE) Exception caught in BufferedNode::getOutput
C:\msys64\home\NAE2\work\src\hark-base\hark-base\src\Collector.cc line 91: Node Collector (type N4HARK9CollectorE) Exception caught in Collector::getOutput

done
initialize
—————————-

Attachments:

1. 

   network.png
   

1. my_sep_rec_offline_2.n

[kohira]

Hi,

First of all, I’d like to thank you.
By amplification, there shows wave in Audacity!

Thank you.
kohira

[kohira]

Hi,

I do not use any files other than ones already found in HARK_recog_3.0.0_practice2.zip.

All that I did are …
1. Extract HARK_recog_3.0.0_practice2.zip
2. cd to HARK_recog_3.0.0_practice2
3. type command shown in console_output.txt

There are 12 wav files generated, but I upload three of these.
I also upload console output log.

Thank you.
kohira

Attachments:

1. result_hark_demo.zip

[kohira]

Hi,

I’ve tried HARK_recog_3.0.0_practice2.zip to get 12 wav files and kaldi_out.txt containing 12 sentences. And accuracies are 80% to 100% for each sentence.

But wav files signals are flat, i.e., no voice or sound.

Could you please tell me how to fix it, or any hint?

Thank you.
kohira

[kohira]

Hi,

My interest is if the separated voice could be used for speech recognition, under the conditions shown below.

1. In a small room
2. 3D position of mouth would be change for several inches during speech.
3. Someone would raise hands during speech, which affects transfer function.

I am very appreciated if I get sample voice separated best.
Or, I would like to know maximum performance of HARK.

Thank you.

[kohira]

Hi,

It is “Transfer Function.”
I created it by calculation for eight microphones and two sources.

Thank you.

[kohira]

Hello,
I made experiment to get result shown below.
1. I have got two stream for two sources which location are fixed.
2. Each has rather large cross talk, but is better than one obtained by just adding two sources waves.
I used TR created by calculation (MIC x 8, fixed source x 2).

Could any body tell me how the TR created by calculation makes it worse than the one by measurement?

Or, is there any sample wave files obtained by HARK’s source separation, which have little cross talk for two or three sources?

Thank you.

[kohira]

The network in Figure 45 is what I needed!
I am trying the network. I will show result in any case.
Thank you.

[kohira]

Oh, sorry for big mistake!
That was “HARK Document Version 3.0.0. (Revision: 9272).”
https://www.hark.jp/document/hark-document-en/subsec-GHDSS.html
Thank you.

[kohira]

Hi,
For source separation under the condition where everything is freeze but for mouths, I think it would be achieved by using certain one node which outputs fix values of directions of sources instead of a set of nodes which are LocalizeMUSIC, SourceTracker … and Delay in Figure 6.56.
But I could not find out the one node yet.
Could you please tell me the node, if you know?
Thank you.

[kohira]

nterakado san,
Thank you, I will try that.

[Author]

Posts