README.md

metadata

title: Audio Translator
emoji: 🔥
colorFrom: pink
colorTo: purple
sdk: gradio
sdk_version: 5.31.0
app_file: app.py
pinned: false
license: apache-2.0
short_description: Audio Translator

🗣️ Audio Translator

[]
[]
[]
[]

---

🚀 Overview

Combine ASR, machine translation, and neural TTS into one seamless audio pipeline—100 % CPU on free-tier HF Spaces.
Upload speech, auto-detect language, translate into English or Spanish, then hear it spoken back.

AI buzzwords:
• Automatic Speech Recognition (ASR) • Whisper Tiny • Neural Machine Translation • GoogleTranslator • Text-to-Speech • gTTS • Multi-modal AI • End-to-End Inference • Real-Time • Edge Deployment

---

✨ Features

🔑 Feature	🔍 Description
🎙️ ASR: Whisper-Tiny	Lightning-fast, on-device speech transcription (all languages)
🌐 Translation	Bidirectional English ↔ Spanish via Deep-Translator
🗣️ Neural TTS	High-quality audio playback via the free Google Translate TTS
⚡ Zero-infra CPU	Runs on 2 vCPU / 16 GB RAM—no GPU or paid APIs needed
🎨 Elegant UI	Intuitive Gradio Blocks—upload, buttons, transcripts, audio
🔧 Fully Modular	Swap models or add logging/analytics with minimal edits

---

🏗️ Architecture & Workflow

1. Audio Upload
   User uploads any .wav or .mp3 clip.
2. ASR
   OpenAI’s whisper-tiny decodes speech into text.
3. MT
   deep-translator’s GoogleTranslator converts text to chosen language.
4. TTS
   gTTS synthesizes the translated text into an .mp3.
5. UI Rendering
   Gradio presents the original transcript, the translation, and an audio player.

---

🛠️ Quick Start (Local Dev)

git clone https://github.com/<YOUR-USERNAME>/audio-translator.git
cd audio-translator
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
python app.py

## Latest Update

- Upgraded Whisper-Tiny model for faster ASR. - May 29, 2025 📝
- Improved translation accuracy for Spanish. ⚡ - May 30, 2025 📝

**Website**: https://ghostainews.com/
**Discord**: https://discord.gg/BfA23aYz