Sfoglia categorie
Esplora
Fiverr Pro
Italiano
$
USD
I will write a python script for speech to text using faster whisper
I'm Sad
Informazioni su questo servizio
Looking for a private, offline AI speech-to-text utility? I will build a custom Python script or a CustomTkinter Desktop App integrated with the cutting-edge Faster-Whisper (Large-v3) model.
Why choose this local AI solution?
- 100% Offline: No API keys, no monthly fees.
- GPU Acceleration: Supports CUDA/cuDNN for fast processing.
- Auto-Extraction: Automatically extracts audio from video files.
Technical Specifications:
1. Inputs:
- Audio: .mp3, .wav, .m4a, .flac, .ogg
- Video: .mp4, .mkv, .avi, .mov, .flv, .wmv
2. Languages: Auto Detect or select 99+ global languages (Vietnamese, English, Chinese, Japanese, Korean, Russian, French, German, Spanish, etc.)
3. Subtitle Export (ExportFormat):
- TXT, SRT, VTT, ASS, XML, TTML
What I offer in my packages:
- Clean CLI Python scripts with installation guides.
- Modern GUI applications with progress bars using CustomTkinter.
- Standalone .exe packages with smart lazy-loading for AI weights.
I deliver clean, well-commented source code and ongoing local support. Let's bring open-source AI straight to your desktop!
Scopri di più su I'm Sad
I'm Sad
Junior Developer Python Fullstack AI
- DaVietnam
- Membro daago 2026
Lingue
Vietnamita, Russo, Inglese
Hi, I'm a 1st-year computer science student. I love turning concepts into practical tools. As a fast learner, I dive into any framework needed to get the job done.
What I can help you with:
- Local AI: Offline speech-to-text (Faster Whisper) & .srt subtitles.
- Desktop Apps: Standalone cross-platform utilities with Tauri & Vanilla JS.
- Frontend: Clean Vanilla HTML/CSS/JS & document readers.
Tech Stack: Python, JavaScript, HTML5/CSS3, Tauri, basic TypeScript.
Experience: Custom Python scripts, Tauri offline tools, and clean frontend layouts for personal clients.
FAQ
Does the app require an internet connection to run?
No, the core transcription process runs 100% offline on your local machine. However, during the very first run, the script can automatically download the required AI model weights from Hugging Face if you haven't downloaded them manually.
Can this application utilize my Nvidia GPU for faster rendering?
Yes! If you purchase the Standard or Premium package, I will configure the Python/CustomTkinter backend to support CUDA and cuDNN, allowing the Faster-Whisper model to leverage your Nvidia graphics card for maximum speed.
Can this application handle video files directly for transcription?
Yes, absolutely! The application supports both audio and video inputs. If you provide a video file (such as .mp4, .mkv, or .mov), the script will automatically handle the audio extraction layer behind the scenes before running the Faster-Whisper AI model.
What subtitle formats can I generate with this utility?
he app is built with a flexible ExportFormat enum. You can export your transcriptions into 6 different formats depending on your needs: TXT (plain text), SRT, VTT, ASS (advanced styled subtitles), XML, and TTML.
