Skip to content

Releases: khalooei/Voxtral-AI-Demo-Local-Interface

Stable-v1.0.0

26 Jul 20:46
@khalooei khalooei
v1.0.0
936ac1e
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.
Compare
Choose a tag to compare
Stable-v1.0.0 Latest
Latest

📢 Release Notes – Voxtral AI Demo Interface v1.0.0 (Stable)

Repository: Voxtral-AI-Demo-Local-Interface

✨ New Features

  • Initial Stable Release of the Voxtral AI Demo Interface.
  • Gradio-based UI for local, interactive inference using Voxtral models.
  • Dual Model Support: Compatible with both Voxtral model versions.
  • Audio Upload & Playback: Supports file input and inline audio playback.
  • Multilingual Transcription: Run high-quality speech-to-text across multiple languages.
  • Integrated Understanding: Enables semantic analysis and long-context audio understanding.

⚙️ Improvements

  • Optimized GPU handling for faster transcription.
  • Responsive UI with metadata inputs and quality control options.
  • Refined UX for seamless testing of audio samples.

🧪 Compatibility

  • ✅ Tested on CUDA-enabled environments with Python 3.10+
  • ✅ Works with both small and large Voxtral variants
  • ✅ Cross-platform (Windows, Linux)

📁 Installation & Usage
Clone and run locally with minimal setup:

git clone https://github.com/khalooei/Voxtral-AI-Demo-Local-Interface.git
cd Voxtral-AI-Demo-Local-Interface
pip install -r requirements.txt
python app.py

📌 Notes

  • Requires a compatible GPU for optimal performance.
  • For best results, use high-quality audio inputs (16kHz recommended).
  • Future updates will include streaming input and speaker diarization support.

🔗 Stay Connected
For issues, feedback, or contributions, visit the GitHub Issues page.

Assets 2
Loading