OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Audio & Voice AI

Deepspeech German

Deepspeech German is a Python project for developing German automatic speech recognition and speech-to-text components with Mozilla DeepSpeech, including documentation for corpus preparation and model training.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Deepspeech German — OpenViST project profile in Audio & Voice AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when building or studying a German speech-recognition component and when you need a repository that documents corpus preparation and model-training stages.

Who it is for

  • Developers adding German transcription to an audio-processing pipeline
  • Researchers or practitioners preparing German speech corpora and training an ASR model

Documented uses

  • Add a German speech-to-text component to an audio-processing workflow.
  • Prepare named German speech corpora for an ASR project.
  • Follow documented steps for training and optimizing a DeepSpeech-based model.
Before you choose

Requirements & limits

The repository is public and lists an Apache-2.0 license.

  • The README targets Mozilla DeepSpeech v0.9.3 and identifies TensorFlow 1.15 and Python 3.7 for the tools.
  • Setup instructions are Linux-focused; other operating systems may require shell and script adjustments.
  • Speech-corpus download URLs may need maintenance.
  • DeepSpeech and KenLM are included as submodules.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare the documented Linux Python environment

Install the Python environment and Linux dependencies described for the project.

Before you start

  • A Linux environment with a shell
  • Python 3.7.9 through pyenv
  • The project's repository and its DeepSpeech and KenLM submodules

Steps

  1. Fetch the DeepSpeech and KenLM submodules with the command below.

    Documented command
    git pull --recurse-submodules

  2. Install pyenv and configure it in the shell as described in the README.

  3. Install Python 3.7.9, create and activate a virtualenv, and install python_requirements.txt.

  4. Install Linux dependencies listed in linux_requirements using the documented apt-get command.

Expected outcome

A Python virtual environment with the listed project requirements installed, plus the documented Linux dependencies.

Guide limitations

  • The instructions focus on Linux; macOS and Windows may require changes.
  • DeepSpeech tools are documented with TensorFlow 1.15, supported up to Python 3.7.
  • The README notes that a patched audiomate version was used.
Documentation 1 ↗
Source-based practical guide

Download the named German speech corpora

Use the project script to request the documented speech-corpus datasets.

Before you start

  • The project Python environment activated
  • The repository's pre-processing/download_speech_corpus.py script

Steps

  1. Activate the project's Python environment.

  2. Run the command below.

    Documented command
    python pre-processing/download_speech_corpus.py --tuda --cv --swc --voxforge --mailabs

Expected outcome

The script is instructed to download the TUDA-De, Common Voice, Spoken Wikipedia, Voxforge and M-AILABS corpora.

Guide limitations

  • The script uses fixed URLs through audiomate that may become outdated.
  • The README says corpus URLs may need updating in the script or to be passed as arguments.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
AASHISHAG/deepspeech-german · Official homepage title · retained in its source language
How we review sourcesSuggest a correction