nodejs package deepspeech crashes (using node v16.0.0)

[solyarisoftware]

My program: deepSpeechTranscriptNative.js, part of a simple project DeepSpeechJs I made to use/test DeepSpeech nodejs bindings APIs, crashes when using latest nodejs version.
Below all details.

My configuration:

(deepspeech-venv) $ inxi -S
System:    Host: giorgio-HP-Laptop-17-by1xxx Kernel: 5.8.0-50-generic x86_64 bits: 64 Desktop: Gnome 3.36.7 
           Distro: Ubuntu 20.04.2 LTS (Focal Fossa) 

(deepspeech-venv) $ deepspeech --version
DeepSpeech  0.9.3

deepspeech-venv) $ npm info deepspeech

deepspeech@0.9.3 | MPL-2.0 | deps: 5 | versions: 97
DeepSpeech NodeJS bindings
https://github.com/mozilla/DeepSpeech/tree/v0.9.3#project-deepspeech

bin: deepspeech

dist
.tarball: https://registry.npmjs.org/deepspeech/-/deepspeech-0.9.3.tgz
.shasum: 675917addc2e2a16fb2bc03718d6a4e6b1bbf650
.integrity: sha512-80yqUWEWgzclY9pjukpq0TdkQEbJ6NzqZ899vsLZfa4YcK35uOWIf+ILK55zQ+Ii/TEJ6Eo62Vc2saedQ/AK6w==
.unpackedSize: 120.3 MB

dependencies:
argparse: 1.0.x      memory-stream: 1.0.x node-pre-gyp: 0.15.x node-wav: 0.0.2      sox-stream: 2.0.x    

maintainers:
- deepspeech-ci <npmjs@deep-speech.org>
- reubenmorais <reuben.morais@gmail.com>
- mozilla-voice-owner <mozilla-voice-package-owner@mozilla.com>

dist-tags:
latest: 0.9.3               prerelease: 0.10.0-alpha.3  

(deepspeech-venv) $ node --version
v16.0.0

CRASH HAPPENS WHEN USING NODEJS V.16

(deepspeech-venv) $ node deepSpeechTranscriptNative.js 
node:internal/modules/cjs/loader:943
  throw err;
  ^

Error: Cannot find module '/home/giorgio/DeepSpeechJs/node_modules/deepspeech/lib/binding/v0.9.3/linux-x64/node-v93/deepspeech.node'
Require stack:
- /home/giorgio/DeepSpeechJs/node_modules/deepspeech/index.js
- /home/giorgio/DeepSpeechJs/deepSpeechTranscriptNative.js
    at Function.Module._resolveFilename (node:internal/modules/cjs/loader:940:15)
    at Function.Module._load (node:internal/modules/cjs/loader:773:27)
    at Module.require (node:internal/modules/cjs/loader:1012:19)
    at require (node:internal/modules/cjs/helpers:93:18)
    at Object.<anonymous> (/home/giorgio/DeepSpeechJs/node_modules/deepspeech/index.js:24:17)
    at Module._compile (node:internal/modules/cjs/loader:1108:14)
    at Object.Module._extensions..js (node:internal/modules/cjs/loader:1137:10)
    at Module.load (node:internal/modules/cjs/loader:988:32)
    at Function.Module._load (node:internal/modules/cjs/loader:828:14)
    at Module.require (node:internal/modules/cjs/loader:1012:19) {
  code: 'MODULE_NOT_FOUND',
  requireStack: [
    '/home/giorgio/DeepSpeechJs/node_modules/deepspeech/index.js',
    '/home/giorgio/DeepSpeechJs/deepSpeechTranscriptNative.js'
  ]
}

NO PROBLEMS WHEN USING NODEJS V.14.16.1

(deepspeech-venv) $ sudo n install 14.16.1
  installing : node-v14.16.1
       mkdir : /usr/local/n/versions/node/14.16.1
       fetch : https://nodejs.org/dist/v14.16.1/node-v14.16.1-linux-x64.tar.xz
   installed : v14.16.1 (with npm 6.14.12)
(deepspeech-venv) $ node --version
v14.16.1

(deepspeech-venv) $ node deepSpeechTranscriptNative.js 

usage: node deepSpeechTranscriptNative [<model pbmm file>] [<model scorer file>] [<audio file>]
using: node deepSpeechTranscriptNative ./models/deepspeech-0.9.3-models.pbmm ./models/deepspeech-0.9.3-models.scorer ./audio/4507-16021-0012.wav

TensorFlow: v2.3.0-6-g23ad988
DeepSpeech: v0.9.3-0-gf2e9c85
2021-05-03 11:32:41.310018: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations:  AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.

pbmm                 : ./models/deepspeech-0.9.3-models.pbmm
scorer               : ./models/deepspeech-0.9.3-models.scorer
elapsed              : 10ms

audio file           : ./audio/4507-16021-0012.wav
transcript           : why should one halt on the way
elapsed              : 3131ms

free model elapsed   : 24ms

Notes:

• it seems to me that the issue is related to NodeJs version 16.0 and previouses?).

• I see that some updates has been done recently in the DeepSpeech project: https://github.com/mozilla/DeepSpeech/tree/master/native_client/javascript

• I have the same issue if I install deepspech package 0.10.0-alpha.3 instead of 0.9.3

• Maybe is the problem related to the deepspeech npm package (last update is 5 months ago) that has to be republished to include updates?

Thanks
Giorgio

[lissyx]

@solyarisoftware There is no support for NodeJS v16 on the uploaded packages. Current packages on master have it but:

• GitHub Actions support only covers linux (amd64, armv7, aarch64), windows (amd64), macOS (amd64)
• Unfortunaly, unlike TaskCluster, we don't have the ability to access unauthenticated to the build artifacts
• There is no support for publishing new releases (Handle tag event on GitHub Actions #3606)
• TaskCluster infra has been completely decomissionned, so producing a 0.9.4 is impossible.

If you are willing to try and help, you are more than welcome. A good first item would be #3606 in your case.

Try and see if you can get deepspeech-tf.tgz or deepspeech-tflite.tgz from the last run: https://github.com/mozilla/DeepSpeech/actions/runs/793741932

[lissyx]

@solyarisoftware The move to GitHub Actions was to make it easier for contributor to actually contribute, please read https://mozilla.github.io/deepspeech-playbook/CONTINUOUS_INTEGRATION.html

[solyarisoftware]

Hi @lissyx
I see your points / thanks for details.
I left the issue open. At least I propose to update this doc https://github.com/mozilla/DeepSpeech/blob/master/native_client/javascript/README.md with a reference to the current issue. Or I'd update the current public npm package version 0.9.3