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

Tevr Asr Tool

Tevr Asr Tool is an MIT-licensed C++ command-line application for offline German speech recognition on Linux x86_64. It processes WAV files with an acoustic AI model and KenLM-based beam-search rescoring, and the project documents operation without a GPU.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tevr Asr Tool — 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 you have German WAV audio and want to run file-based transcription on Linux, including a CPU-only machine.

Who it is for

  • Linux users who need to transcribe German WAV recordings locally.
  • Developers exploring or adapting a compact C++ speech-recognition workflow or its Hugging Face model.

Documented uses

  • Convert a German WAV recording into text from a Linux command line.
  • Keep audio transcription offline rather than sending recordings to a cloud service.
  • Explore a documented model fine-tuning route for another language dataset.
Before you choose

Requirements & limits

The project README describes the tool as free and MIT-licensed, including commercial use, and states that it comes with no warranty.

  • The documented input is a WAV file, and the packaged target is Linux x86_64.
  • The README says a GPU is not required.
  • The processing flow uses an acoustic model followed by KenLM language-model rescoring.
  • The reported 3.64% word error rate and first-place Common Voice German ranking refer to the project's August 2022 benchmark claim.
  • The README describes real-time GPU transcription as planned, separate from the documented file-based CLI tool.

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

Install and transcribe a WAV file on Debian or Ubuntu

Install the documented Linux x86_64 package and pass a WAV file to the command-line transcription tool.

Before you start

  • Debian or Ubuntu on Linux x86_64.
  • The five numbered ZIP parts for the v1.0.0 package, plus wget, unzip and dpkg.
  • A WAV file to transcribe.

Steps

  1. Download the five numbered ZIP parts with the wget commands in the project README.

  2. Join the parts into one ZIP file and extract it using the documented cat and unzip commands.

  3. Install the extracted Debian package with dpkg.

  4. Run tevr_asr_tool --target_file=test_audio.wav 2>log.txt, replacing the example filename with the WAV file to transcribe.

Expected outcome

The command displays a transcription; diagnostics and progress are redirected to log.txt.

Guide limitations

  • The documented package and command target Linux x86_64 and WAV input.
  • The README’s sample transcription is an example, not an accuracy guarantee for other audio.
Documentation 1 ↗
Evidence and transparency

Sources & review

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