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

Whisper

Whisper is an open-source general-purpose speech recognition model and software package. Its Transformer model handles multilingual transcription, speech translation, spoken-language identification and voice activity detection, with command-line and Python usage documented in its repository.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Whisper — 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 Whisper when you need a self-hosted model for audio transcription or language-aware speech processing and can choose a model based on available memory, speed and task requirements.

Who it is for

  • Developers who want to run speech recognition locally or in their own environment.
  • People building workflows for multilingual transcription, English translation of non-English speech, or spoken-language identification.

Documented uses

  • Transcribe speech from audio files.
  • Translate non-English speech into English using a multilingual model.
  • Identify the language spoken in an audio segment through the Python interface.
Before you choose

Requirements & limits

The repository releases Whisper's code and model weights under the MIT License. It documents installation with pip and requires ffmpeg.

  • Model size trades off speed and memory requirements; the README lists approximate VRAM from about 1 GB for tiny and base to about 10 GB for large.
  • Turbo is intended for fast transcription, but does not perform translation even when the translate task is specified.
  • For translation into English, choose a multilingual model; the README recommends medium or large for the best translation results.
  • Performance varies by language, speaking speed and available hardware.

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 Whisper for command-line use

Set up the Whisper package and its documented audio dependency.

Before you start

  • Python; the repository says the codebase is expected to be compatible with Python 3.8–3.11 and recent PyTorch versions.
  • The ffmpeg command-line tool.

Steps

  1. Install ffmpeg using the package manager for your operating system.

  2. Install Whisper with the command below.

    Documented command
    pip install -U openai-whisper

Expected outcome

Whisper and its command-line tool are installed; the repository documents `whisper --help` for viewing available options.

Guide limitations

  • Rust may be needed if tiktoken does not provide a pre-built wheel for the platform.
  • This recipe covers setup and option discovery, not a transcription run.
Documentation 1 ↗
Source-based practical guide

Transcribe an audio file from the command line

Use the documented CLI form to transcribe an audio file.

Before you start

  • Whisper installed and ffmpeg available.
  • An audio file accessible to the command line.

Steps

  1. Choose a model; the documented default is turbo.

  2. Run whisper audio.flac audio.mp3 audio.wav --model turbo, replacing the example filenames with the audio files to process.

Expected outcome

Whisper processes the supplied audio files for transcription.

Guide limitations

  • The documented command is for transcription; turbo is not trained for speech translation.
  • Speed and resource needs vary by model, language and hardware.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
openai/whisper · Official homepage title · retained in its source language
How we review sourcesSuggest a correction