# Using Faster Whisper on macOS

**1. Open the terminal and activate the environment**

{% code overflow="wrap" %}

```
source whisper-env/bin/activate
```

{% endcode %}

**2. Open the script**

{% code overflow="wrap" %}

```
nano faster_transcribe.py
```

{% endcode %}

**3.1. Edit the script if transcription from audio is required** (model size on Line 5 and/or audio file path and name on Line 8)

{% code lineNumbers="true" %}

```
from faster_whisper import WhisperModel
import os

# Load the model
model = WhisperModel(“large-v3”, device="auto", compute_type="int8")

# Path to your audio file
audio_path = "/Users/nombreUsuario/Desktop/Audio.m4a"

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

# Print language
print("Detected language:", info.language)

# Build transcript text
transcript_lines = []
for segment in segments:
    line = f"[{segment.start:.2f} - {segment.end:.2f}] {segment.text}"
    print(line)
    transcript_lines.append(line)
    
# Save to a .txt file in the same folder as the audio file
base_filename = os.path.splitext(os.path.basename(audio_path))[0]
output_path = os.path.join(os.path.dirname(audio_path), f"{base_filename}_transcript.txt")

with open(output_path, "w", encoding="utf-8") as f:
    f.write("\n".join(transcript_lines))
print(f"\n✅ Transcript saved to: {output_path}")
```

{% endcode %}

**3.2. Edit the script if transcription from video is required** (model size on Line 6 and/or video file path and name on Line 9)

{% code lineNumbers="true" %}

```
from faster_whisper import WhisperModel
import subprocess
import os

# Load the model
model = WhisperModel("large-v3", device="auto", compute_type="int8")

# Path to your video file
video_path = "/Users/nombreUsuario/Desktop/Video.mp4"

# Extract audio as a temporary file (.wav for compatibility)
audio_path = os.path.splitext(video_path)[0] + "_temp_audio.wav"
subprocess.run([
    "ffmpeg", "-i", video_path,
    "-vn",              # no video
    "-acodec", "pcm_s16le",  # compatible audio format
    "-ar", "16000",     # sample rate recommended by Whisper
    "-ac", "1",         # single channel (mono)
    audio_path,
    "-y"                # overwrite if it already exists
], check=True)

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

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

# Build transcript text
transcript_lines = []
for segment in segments:
    line = f"[{segment.start:.2f} - {segment.end:.2f}] {segment.text}"
    print(line)
    transcript_lines.append(line)

# Save transcript as .txt next to the original video
base_filename = os.path.splitext(os.path.basename(video_path))[0]
output_path = os.path.join(os.path.dirname(video_path), f"{base_filename}_transcript.txt")
with open(output_path, "w", encoding="utf-8") as f:
    f.write("\n".join(transcript_lines))

print(f"\n✅ Transcript saved to: {output_path}")

# (Optional) Delete temporary audio file
os.remove(audio_path)
```

{% endcode %}

**4. Save**

* Press CTRL + O, then Enter
* Press CTRL + X to exit Nano

**5. Run the script**

```
python faster_transcribe.py
```

**6. Output**

* Transcription will appear in the terminal
* A .txt file will be saved on the 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-macos.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.