How to fix the error 'u/3�', should be 'TFL3'?


so after my first installation following the guide at I tried to use it with the provided command:
deepspeech --model deepspeech-0.9.3-models.pbmm --scorer deepspeech-0.9.3-models.scorer --audio audio/2830-3980-0043.wav, but if I do that, I get an error:

Loading model from file deepspeech-0.9.3-models.pbmm
TensorFlow: v2.3.0-6-g23ad988
DeepSpeech: v0.9.3-0-gf2e9c85
ERROR: Model provided has model identifier 'u/3�', should be 'TFL3'

Error at reading model file deepspeech-0.9.3-models.pbmm
Traceback (most recent call last):
  File "/home/pi/tmp/deepspeech-venv/bin/deepspeech", line 8, in <module>
  File "/home/pi/tmp/deepspeech-venv/lib/python3.7/site-packages/deepspeech/", line 119, in main
    ds = Model(args.model)
  File "/home/pi/tmp/deepspeech-venv/lib/python3.7/site-packages/deepspeech/", line 38, in __init__
    raise RuntimeError("CreateModel failed with '{}' (0x{:X})".format(deepspeech.impl.ErrorCodeToErrorMessage(status),status))
RuntimeError: CreateModel failed with 'Failed to initialize memory mapped model.' (0x3000)`

What’s the problem and how can I fix that?
I couldn’t find a solution on Google or at this forum, so I decided to create a new topic. If there is any explanation online that I just didn’t see, I’d appreciate it if someone posted it here.

I am pretty much a complete beginner with Rasperrys / Linux so any advice is really helpful. If you think that DeepSpeech is too much of a challenge for a beginner, please say it.

Thank you very much.

No worries, it can be complicated. But just today somebody linked a good post on how to run on Raspberry.

You used the desktop model .pbmm which is too big for such a small device. Use the .tflite version of the model instead and it should work.

Good luck and let us know if you have any other problems

1 Like

It works now!
Thank you for your advice.

Hello…I have the same problem…Almost exact error syntax only that I’m running the code in Windows 10, python3.8 in PyCharm. Help me out.

@HarleyDuncan please don’t spam our old threads, you just opened a new thread. Just delete this post.

@HarleyDuncan Select engine “ds” intead of “stt”

No, this is technically not the correct answer. You need to use TFLite runtime. As a matter of fact, STT project ditched using the protobuf implem and since several releases they only used TFLite, so yes, use stt works, but not because DS vs STT.