Whisper

Whisper (OpenAI) is the industry standard for Speech-to-Text. It supports 99 languages and translation. V3 (large-v3) is the current state of the art.

When to Use

• Speech-to-Text Audio Transcription: Transcribing multi-lingual audio recordings into punctuated, accurate text.
• Audio Translation to English: Automatically translating non-English speech directly into English transcripts.
• Word-Level Subtitles & Timestamps: Generating SRT and VTT subtitles with word-level alignment for video editors.
• High-Throughput Inference with Faster-Whisper: Running CTranslate2-accelerated inference with 4x speedup and lower VRAM.

Quick Start

import whisper

# Load pre-trained model and transcribe audio file
model = whisper.load_model("base")
result = model.transcribe("interview.mp3")