# Using Faster Whisper on Windows

**0. Create a script file in a text editor**, save it as “transcribe.py”, selecting "All Files" as the file type when saving — not .txt.

**1.1. Edit the script if transcription from audio is needed.** Modify “transcribe.py” by adjusting the model size (Line 7) and/or the audio file path and name (Line 4).

{% code title="transcribe.py" lineNumbers="true" %}

```
from faster_whisper import WhisperModel

# Path to the audio file
audio_path = r"C:\Users\User\Music\Audio.ogg"

# Model size (you can change to "base", "medium", "large-v2", etc.)
model = WhisperModel("small")

# Transcribe the audio
segments, info = model.transcribe(audio_path)

# Display the detected language and the text
print("Detected language:", info.language)

for segment in segments:
    print(f"[{segment.start:.2f}s -> {segment.end:.2f}s] {segment.text}")
```

{% endcode %}

**1.2. Edit the script if transcription from video is needed.** Modify “transcribe.py” by adjusting the model size (Line 37) and/or the video file path and name (Line 8).

{% code title="transcribe.py" lineNumbers="true" %}

```
import os
import subprocess
from faster_whisper import WhisperModel
from pathlib import Path

# === Configuration ===
# Path to the video
video_path = r"C:\Users\User\Videos\video.mp4"

# Temporary folder for the extracted audio
temp_audio_path = "temp_audio.wav"

# Current user's Desktop folder
desktop_folder = str(Path.home() / "Desktop")
output_text_file = os.path.join(desktop_folder, "transcription.txt")

# === Step 1: Extract audio from video using ffmpeg ===
print(" Extracting audio from video...")
ffmpeg_command = [
    "ffmpeg", "-y",  # -y to overwrite without asking
    "-i", video_path,
    "-vn",           # No video
    "-acodec", "pcm_s16le",  # Uncompressed audio (WAV)
    "-ar", "16000",  # Sample rate
    "-ac", "1",      # Mono
    temp_audio_path
]

try:
    subprocess.run(ffmpeg_command, check=True)
except subprocess.CalledProcessError:
    print(" Error extracting audio with ffmpeg.")
    exit(1)

# === Step 2: Transcribe with Faster Whisper (you can change to "base", "medium", "large-v2", etc.) ===
print(" Transcribing audio...")
model = WhisperModel("large-v2")

segments, info = model.transcribe(temp_audio_path)

# === Step 3: Save transcription ===
print("Saving transcription...")
with open(output_text_file, "w", encoding="utf-8") as f:
    f.write(f"Detected language: {info.language}\n\n")
    for segment in segments:
        f.write(f"[{segment.start:.2f}s -> {segment.end:.2f}s] {segment.text}\n")

print(f"Transcription saved to: {output_text_file}")

# === Step 4: Remove temporary audio file ===
os.remove(temp_audio_path)
print(" Temporary audio file deleted.")
```

{% endcode %}

**2. Open terminal in the script folder**

I. Open File Explorer and navigate to the folder where you saved transcribe.py.

II. Right-click on a blank space inside the folder (not on a file).

III. Select “Open in Terminal” (on older versions of Windows, it may appear as “Open PowerShell window here”).

**3. Run the script**

```
python transcribe.py
```

**4. Output**

* The transcription will appear directly in the terminal.
* A .txt file named transcription.txt will be saved on your Desktop.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://dojo.lkmx.io/ingles/faster-whisper/using-faster-whisper-on-windows.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.