Audio Toolbox transcribe not working even with toolbox and license

See the following command line output:

Error using Wav2VecSpeechClient/transcribe

Audio Toolbox is required for wav2vec 2.0. Install Audio Toolbox and activate a license.

Error in speech2text

Error in abdulazd_3W03_Lab8_script (line 57)

transcript = speech2text(transcriber,y,fs);

>> ver

-----------------------------------------------------------------------------------------------------

MATLAB Version: 9.14.0.2337262 (R2023a) Update 5

MATLAB License Number: 40924515

Operating System: Microsoft Windows 11 Famille Version 10.0 (Build 22621)

Java Version: Java 1.8.0_202-b08 with Oracle Corporation Java HotSpot(TM) 64-Bit Server VM mixed mode

-----------------------------------------------------------------------------------------------------

MATLAB Version 9.14 (R2023a)

Audio Toolbox Version 3.4 (R2023a)

Curve Fitting Toolbox Version 3.9 (R2023a)

DSP System Toolbox Version 9.16 (R2023a)

Image Processing Toolbox Version 11.7 (R2023a)

Signal Processing Toolbox Version 9.2 (R2023a)

Symbolic Math Toolbox Version 9.3 (R2023a)

>> license('inuse')

audio_system_toolbox

matlab

neural_network_toolbox

signal_blocks

signal_toolbox

I'm confused why the transcribe method isn't working. I have a license and the toolbox downloaded, according to the ver and license commands, right? I am able to run transcriber = speechClient("wav2vec2.0"); without any issues, so I think I downloaded the toolbox properly.

Answer 1

[Steven Lord]

Looking at the documentation page for the speech2text function, I see a Note in the Decription section that looks relevant.

"To use speech2text with the third-party speech services, you must download the extended Audio Toolbox™ functionality from File Exchange. The File Exchange submission includes a tutorial to get started with the third-party services.

Using wav2vec 2.0 requires Deep Learning Toolbox™ and installing the pretrained model."

The error message you received does seem incorrect to me. Based on that Note and the output of ver you posted, it probably should have indicated you don't have Deep Learning Toolbox installed instead of Audio Toolbox. I've reported this to the development staff.

[Dabeer Abdul-Azeez]

You're absolutely correct! Installing the Deep Learning Toolbox allowed it work. I didn't need to use the File Exchange link you sent, though, just installed the Audio Toolbox and the Deep Learning Toolbox.

Thanks a million :)