Managing FastAPI Projects with Poetry: A Step-by-Step Guide

-
Image
This guide will walk you through how to use Poetry to manage dependencies and structure your project in FastAPI . It includes detailed explanations of Poetry's core concepts and commands to help prevent issues that can arise during team collaboration and deployment. 1. What is Poetry? Poetry is a dependency management and packaging tool for Python. It goes beyond simply installing libraries by allowing you to clearly declare the dependencies your project needs and ensuring that all developers on the project have the same library versions. Clarity in Dependency Management : Explicitly manage your project's basic information and required libraries through the pyproject.toml file. Reproducible Builds : By locking all dependency versions in the poetry.lock file, it fundamentally prevents "it works on my machine" problems. Integrated Development Environment : It automatically creates and manages isolated virtual environments for each project and handles mo...
Post a Comment

Speech-to-Text and Speaker Diarization with Whisper and pyannote.audio 3.1 on Ubuntu 24.04

-
open ai whisper

If you want to generate subtitles with speaker labels from an audio file using Whisper and pyannote.audio on Ubuntu 24.04, this blog post walks you through the full setup process.

Step 1: Python 3.10 Installation (Manual)

Ubuntu 24.04 comes with Python 3.12 by default, but pyannote.audio 3.1.1 was trained with torch==1.13.1 and CUDA 11.7, which supports Python 3.10. So I need to install Python 3.10:

Step 2: Create and Activate a Python Virtual Environment

stt@GU502DU:~$ python3.10 -m venv whisper_env
stt@GU502DU:~$ source whisper_env/bin/activate

Step 3: Install torch 2.5.1 and cuda 12.1

My laptop is equipped with an NVIDIA GeForce GTX 1660, so I chose to install and use CUDA. Since the GPU driver reports CUDA Version 12.8, I installed the cu121 build of PyTorch, which is compatible with it.

(whisper_env) stt@GU502DU:~$ pip install torch==2.5.1+cu121 torchaudio==2.2.1+cu121 \
  --extra-index-url https://download.pytorch.org/whl/cu121

Note on CUDA Compatibility

To ensure compatibility between your installed CUDA version and the PyTorch build, run the following command:

(diarization_env) stt@GU502DU:~$ nvidia-smi
+------------------------------------------------------------------------+
| NVIDIA-SMI 570.133.07  Driver Version: 570.133.07  CUDA Version: 12.8  |
+------------------------------------------------------------------------+

Make sure that the CUDA version reported here (e.g., CUDA Version: 12.8) is greater than or equal to the CUDA version used by the installed PyTorch build (in this case, cu121 = CUDA 12.1).

Step 4: pyannote.audio and dependencies:

(whisper_env) stt@GU502DU:~$ pip install pyannote.audio==3.1.1 numpy scipy librosa huggingface_hub

Step 5: Install Whisper (with GPU support)

(whisper_env) stt@GU502DU:~$ pip install git+https://github.com/openai/whisper.git --upgrade --no-deps

I use the --no-deps option to prevent pip from overriding already-installed packages like torch.

Step 6: Hugging Face Access Token

I create a Hugging Face access token from hf.co/settings/tokens and use it in my code:

from huggingface_hub import login
login("hf_your_token_here")

Step 7: Accept User Conditions for the Following Models

I visit the following URLs and click “Agree” to accept the terms:

Step 8: Run the following code (Whisper + pyannote.audio)

(whisper_env) stt@GU502DU:~$ python3 speaker_test4.py
from pyannote.audio import Pipeline
import whisper
import json

# Diarization
pipeline = Pipeline.from_pretrained(
    "pyannote/speaker-diarization-3.1",
    use_auth_token="hf_your_token"
)
diarization = pipeline("test.wav")

# STT
model = whisper.load_model("base")
whisper_result = model.transcribe("test.wav")

# Merge
def get_speaker(start_time, end_time, diarization):
    for turn, _, speaker in diarization.itertracks(yield_label=True):
        overlap = max(0, min(end_time, turn.end) - max(start_time, turn.start))
        if overlap > (end_time - start_time) * 0.5:
            return speaker
    return "Unknown"

merged = []
for seg in whisper_result["segments"]:
    speaker = get_speaker(seg["start"], seg["end"], diarization)
    merged.append({
        "speaker": speaker,
        "start": seg["start"],
        "end": seg["end"],
        "text": seg["text"]
    })

with open("test_stt_merged.json", "w") as f:
    json.dump(merged, f, indent=2)

Output

This generates test_stt_merged.json that combines speaker info and transcribed text.

{
  "speaker": "SPEAKER_00",
  "start": 11.200000000000001,
  "end": 19.36,
  "text": " you this morning? Um, I just had some, um, diary for the last three days. Um, and it's"
},
{
  "speaker": "SPEAKER_00",
  "start": 19.36,
  "end": 24.16,
  "text": " been affecting me. I need to stay close to the toilet and, um, yeah, it's been affecting"
},
Location: Calgary, AB, Canada

Comments

Post a Comment

Popular posts from this blog

Resolving Key Exchange Failure When Connecting with SecureCRT to OpenSSH

-
Image
Recently, after installing OpenSSH on my Ubuntu 22.04 laptop, I encountered a challenge while trying to establish a connection using SecureCRT. An error message appeared: Key exchange failed. No compatible key-exchange method. The server supports these methods:   curve25519-sha256,curve25519-sha256@libssh.org,ecdh-sha2-nistp256,ecdh-sha2-nistp384,ecdh-sha2-nistp521,sntrup761x25519-sha512@openssh.com,diffie-hellman-group-exchange-sha256,diffie-hellman-group16-sha512,diffie-hellman-group18-sha512,diffie-hellman-group14-sha256,kex-strict-s-v00@openssh.com The following key-exchange method(s) are supported but not currently allowed  for this session:   curve25519-sha256,curve25519-sha256@libssh.org,ecdh-sha2-nistp256,ecdh-sha2-nistp384,ecdh-sha2-nistp521,sntrup761x25519-sha512@openssh.com,diffie-hellman-group-exchange-sha256,diffie-hellman-group16-sha512,diffie-hellman-group18-sha512,diffie-hellman-group14-sha256,kex-strict-s-v00@openssh.com,curve25519-sha256 Key-exchange m...
Read more

SecureCRT] How to Back Up and Restore SecureCRT Settings on Windows

-
Image
If you're a frequent user of SecureCRT , backing up your session settings and restoring them later is crucial—especially when migrating to a new PC or recovering from a crash. In this guide, you’ll learn how to manually back up and restore SecureCRT settings on Windows , ensuring you never lose your configurations again. Does SecureCRT Have a Built-in Export/Import Feature? Yes, SecureCRT provides built-in export and import features in the Tools menu. However, these features may not capture all configuration details such as: Backup scripts Custom login scripts Certain session-specific files Therefore, if you want to completely clone your SecureCRT environment—including every session and customization—the safest option is to manually back up the entire Config folder. What Is the SecureCRT Config Folder? SecureCRT stores your personal settings (sessions, host keys, global options, etc.) in a folder located at: C:\Users\<Your username>\AppData\Roaming\VanDyke\Config Th...
Read more

How to Set Up Vaultwarden (Bitwarden) on Synology NAS (Best Free Alternative to LastPass)

-
Image
Have you ever felt overwhelmed managing passwords for all your websites? I used to use the same password across multiple sites , which wasn’t secure. To manage them better, I started using LastPass for years. However, in 2019, they changed their pricing policy — now it’s free only for one device for 30 days, and you have to pay to use it across multiple devices . If you're looking for a powerful and free password manager, Vaultwarden is one of the best alternatives to LastPass. What is Vaultwarden? Vaultwarden (formerly known as Bitwarden_RS) is a lightweight, open-source implementation of the Bitwarden server API written in Rust . It is fully compatible with official Bitwarden clients and is ideal for low-resource environments. Example: On a Synology DS916+, Vaultwarden uses 0% CPU and only 64MB of memory while running. My Setup NAS Model : Synology DS916+ DSM Version : DSM 7.1-42661 Update 1 Docker Image : vaultwarden/server Ins...
Read more