Bump libdeepspeech from 0.7.1 to 0.7.4

[dependabot-preview[bot]]

Bumps libdeepspeech from 0.7.1 to 0.7.4.

Release notes

Sourced from libdeepspeech's releases.

DeepSpeech 0.7.4

General

This is the 0.7.4 release of Deep Speech, an open speech-to-text engine. In accord with semantic versioning, this version is not backwards compatible with version 0.6.1 or earlier versions. This is a bugfix release and retains compatibility with the 0.7.0 models. All model files included here are identical to the ones in the 0.7.0 release. As with previous releases, this release includes the source code:

v0.7.4.tar.gz

and the acoustic models:

deepspeech-0.7.4-models.pbmm deepspeech-0.7.4-models.tflite.

The model with the ".pbmm" extension is memory mapped and thus memory efficient and fast to load. The model with the ".tflite" extension is converted to use TFLite, has post-training quantization enabled, and is more suitable for resource constrained environments.

The acoustic models were trained on American English and the pbmm model achieves an 5.97% word error rate on the LibriSpeech clean test corpus.

In addition we release the scorer:

deepspeech-0.7.4-models.scorer

which takes the place of the language model and trie in older releases.

We also include example audio files:

audio-0.7.4.tar.gz

which can be used to test the engine, and checkpoint files:

deepspeech-0.7.4-checkpoint.tar.gz

which can be used as the basis for further fine-tuning.

Notable changes from the previous release

• Fix csv.DictWriter configuration on Windows in some importers (#3045)
• Reduce number of users of VERSION and GRAPH_VERSION symlinks to fix issues on Windows (#3043)
• Fix bug in ds_ctcdecoder SWIG definition which was causing wrapper objects to be leaked (#3049)
• Add support for read-only validation metrics (not affecting best validation checkpoint logic) (#3051)
• Fix some importers to report total imported audio duration alongside total input audio duration (#3054)
• Separate Dockerfile into one for training and one for building native client related tools (#3060)
• Add list of supported platforms to ReadTheDocs (#3065)
• Added third-party bindings for the Nim language (#3076)
• Avoid reinstalling TensorFlow package from PyPI when using Docker bases that already come with it (#3072)
• Refactor artifact caching mechanism in CI (#3069)

Training Regimen + Hyperparameters for fine-tuning

The hyperparameters used to train the model are useful for fine tuning. Thus, we document them here along with the training regimen, hardware used (a server with 8 Quadro RTX 6000 GPUs each with 24GB of VRAM), and our use of cuDNN RNN.

In contrast to previous releases, training for this release occurred in several phases each phase with a lower learning rate than the phase before it.

Changelog

Sourced from libdeepspeech's changelog.

Commits

• fcd9563 Merge pull request #3085 from mozilla/new-version-074
• 4c6245d Merge pull request #3055 from tilmankamp/augext
• 5edc1cf Bump VERSION to 0.7.4
• bc31eb4 Fix usage of ARG instead of ENV in Dockerfile.train
• 188a6f2 Merge pull request #3080 from mozilla/install-instructions
• 12a24b8 Merge pull request #3083 from DanBmh/fix_docker
• 3f8033e Add dependencies for new audio augmentation flags. Fixes #3082.
• 6ccbbed Remove --force-reinstall from training code install
• b7fa0ad Merge pull request #3072 from lissyx/docker-train
• 07c8dae Update setup.py
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot merge` will merge this PR after your CI passes on it - `@dependabot squash and merge` will squash and merge this PR after your CI passes on it - `@dependabot cancel merge` will cancel a previously requested merge and block automerging - `@dependabot reopen` will reopen this PR if it is closed - `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) - `@dependabot use these labels` will set the current labels as the default for future PRs for this repo and language - `@dependabot use these reviewers` will set the current reviewers as the default for future PRs for this repo and language - `@dependabot use these assignees` will set the current assignees as the default for future PRs for this repo and language - `@dependabot use this milestone` will set the current milestone as the default for future PRs for this repo and language - `@dependabot badge me` will comment on this PR with code to add a "Dependabot enabled" badge to your readme Additionally, you can set the following in your Dependabot [dashboard](https://app.dependabot.com): - Update frequency (including time of day and day of week) - Pull request limits (per update run and/or open at any time) - Out-of-range updates (receive only lockfile updates, if desired) - Security updates (receive only security updates, if desired)