Whisper Speech to Text: Free Local Transcription Solution

Recording a one-hour meeting, then spending two hours organizing notes. Is there a way to automatically transcribe the recording into text? Whisper can do it, and it's completely free, runs locally, and ensures privacy and security.

💡 What You Will Learn

Recording a one-hour meeting, then spending two hours organizing notes. Is there a way to automatically transcribe the recording into text? Whisper can do it, and it's completely free, runs locally, a

WhisperWhat Is

# Installation/Setup
pip install openai-whisper

# 
whisper .mp3 --model small --language Chinese

|:----|:---:|:----:|:-----:|:--------| | tiny | 39MB |||| | base | 74MB |||| | small | 244MB |||| | medium | 769MB |||| | large | 1.5GB ||||

import whisper

model = whisper.load_model("medium")
result = model.transcribe(".mp3", language="zh")

# 
print(result["text"])

# 
for segment in result["segments"]:
    start = segment["start"]
    end = segment["end"]
    text = segment["text"]
    print(f"[{start:.1f}s - {end:.1f}s] {text}")

pip install whisperx

whisperx .mp3 --model medium --language zh

[ 00:00:05] 
[ 00:00:12] 
[ 00:00:30] 

|:----|:----|:----|:-----:| |:----|:-------:|:----| | tiny | 1GB || | small | 2GB || | medium | 5GB || | large | 10GB ||