Items with no label
3335 Discussions

Audio Recognition Error (Audio File MP3)

hsait6
Beginner
1,874 Views

Hi All

I'm using RealSense SDK 2016 R2.

I installed intel_rs_dcm_sr300_3.3.27.5718 and intel_rs_sdk_offline_package_10.0.26.0396

voice recognition works properly when Audio Source is SR300.

However, when I specified Audio File as Audio Source, PXC_STATUS_PARAM_UNSUPPORTED error occurred.

Example 146 in the following document, "The SDK audio source supports reading from an audio file." is written

https://software.intel.com/sites/landingpage/realsense/camera-sdk/v1.1/documentation/html/index.html?doc_speech_speech_recognition_procedure.html Intel® RealSense™ SDK 2016 R2 Documentation

But, The return value of SetDevice function is PXC_STATUS_PARAM_UNSUPPORTED and it does not work properly.

Windows 10 Home

Best Regards

0 Kudos
4 Replies
idata
Employee
312 Views

Hi Hiroshi.Saito,

 

 

Thanks for your interest in the Intel Realsense Platform.

 

 

We are currently investigating your issue, as soon as we have some useful information we will contact you back.

 

 

Have a nice day!

 

 

Best Regards,

 

-Jose.
0 Kudos
idata
Employee
312 Views

Hi Hiroshi.Saito,

 

 

I appreciate your patience.

 

 

Looking at the SDK documentation https://software.intel.com/sites/landingpage/realsense/camera-sdk/v1.1/documentation/html/index.html?doc_speech_audio_data.html Intel® RealSense™ SDK 2016 R2 Documentation it seems that the SDK supports audio formats as listed in the AudioFormat enumerator. See the AudioFormat enumerator description for details: https://software.intel.com/sites/landingpage/realsense/camera-sdk/v1.1/documentation/html/index.html?audioformat_pxcaudio.html Intel® RealSense™ SDK 2016 R2 Documentation .

 

 

As you can see if you want to use a MP3 file you will need to decode or transcode first.

 

 

I hope you find this information useful, have a nice day!

 

 

Best Regards,

 

-Jose.

 

0 Kudos
hsait6
Beginner
312 Views

Hi Jose,

Thank you for your reply.

But I didn't know how to deal with your suggestion.

So, I tested '.wav' audio file (convert from MP3 to WAV)

As a result, I confirmed that speech recognition works properly in the procedure of Example 146 to 150.

Audio file (14 minutes 33 seconds) was converted to text in 1 minute 50 seconds.

(Speech recognition engine is 'LANGUAGE_US_ENGLISH')

Thus, I decided to use wav file when I use audio files.

Best Regards,

0 Kudos
idata
Employee
312 Views

Hi Hiroshi.Saito,

 

 

I'm glad to hear that you solve your issue! Thanks for sharing your results, since it may benefit other users in the future.

 

 

If you have any more questions, please don't hesitate to contact us.

 

 

Have a nice day!

 

 

Best Regards,

 

-Jose.
0 Kudos
Reply