---
title: 'Multilingual TTS: Amharic, Oromo, Somali, English & More'
emoji: 🌍
colorFrom: gray
colorTo: purple
sdk: gradio
sdk_version: 5.42.0
app_file: app.py
pinned: false
license: mit
short_description: Text-to-Speech for Amharic, Oromo, English, and more.
---

# Multilingual Text-to-Speech Converter

<div align="center">
  <img src="httpshttps://raw.githubusercontent.com/snackshell/eng-amh-tts/main/assets/demo.png" width="800" alt="Multilingual TTS Interface">
  <br>
  <em>Natural sounding speech synthesis for a wide variety of languages.</em>
</div>

[![Open in Spaces](https://img.shields.io/badge/🤗-Try%20Live-blue)](https://huggingface.co/spaces/snackshell/selam-tts)

## Key Features 🌟
- **Multilingual Support**: Seamlessly generate speech for Amharic, Oromo, Somali, Tigrinya, English, Arabic, French, German, Italian, Japanese, Korean, and Chinese.
- **Multiple Natural Voices**: Each language comes with distinct male and female speaker options.
- **Smart Fallbacks**: Unsupported languages like Tigrinya and Oromo use the closest available high-quality voice for pronunciation.
- **Real-time Generation**: Fast response time for most requests.
- **Modern UI**: Clean and easy-to-use interface powered by Gradio.

## How to Use 🎤
1. **Select Language**: Choose your desired language from the dropdown.
2. **Choose Voice**: Pick your preferred speaker.
3. **Enter Text**: Type or paste your text.
4. **Generate**: Click the "Generate Audio" button.
5. **Play**: Listen directly in your browser.

## Supported Voices 🎧
| Language | Speaker | Gender | Voice ID / Fallback |
| :--- | :--- | :--- | :--- |
| Amharic 🇪🇹 | Mekdes | Female | `am-ET-MekdesNeural` |
| Amharic 🇪🇹 | Ameha | Male | `am-ET-AmehaNeural` |
| Oromo 🇪🇹 | Zuri | Female | `sw-KE-ZuriNeural` (Swahili) |
| Oromo 🇪🇹 | Rafiki | Male | `sw-KE-RafikiNeural` (Swahili) |
| Somali 🇸🇴 | Ubax | Female | `so-SO-UbaxNeural` |
| Somali 🇸🇴 | Muuse | Male | `so-SO-MuuseNeural` |
| Tigrinya 🇪🇷 | Lulia | Female | `am-ET-MekdesNeural` (Amharic)|
| Tigrinya 🇪🇷 | Birhane | Male | `am-ET-AmehaNeural` (Amharic)|
| English 🇬🇧 | Sonia | Female | `en-GB-SoniaNeural` |
| English 🇬🇧 | Ryan | Male | `en-GB-RyanNeural` |
| Arabic 🇸🇦 | Zariyah | Female | `ar-SA-ZariyahNeural` |
| Arabic 🇸🇦 | Hamed | Male | `ar-SA-HamedNeural` |
| French 🇫🇷 | Denise | Female | `fr-FR-DeniseNeural` |
| French 🇫🇷 | Henri | Male | `fr-FR-HenriNeural` |
| German 🇩🇪 | Katja | Female | `de-DE-KatjaNeural` |
| German 🇩🇪 | Conrad | Male | `de-DE-ConradNeural` |
| Italian 🇮🇹 | Elsa | Female | `it-IT-ElsaNeural` |
| Italian 🇮🇹 | Diego | Male | `it-IT-DiegoNeural` |
| Japanese 🇯🇵 | Nanami | Female | `ja-JP-NanamiNeural` |
| Japanese 🇯🇵 | Keita | Male | `ja-JP-KeitaNeural` |
| Korean 🇰🇷 | Sun-Hi | Female | `ko-KR-SunHiNeural` |
| Korean 🇰🇷 | InJoon | Male | `ko-KR-InJoonNeural` |
| Chinese (Simp.) 🇨🇳 | Xiaoxiao | Female | `zh-CN-XiaoxiaoNeural` |
| Chinese (Simp.) 🇨🇳 | Yunxi | Male | `zh-CN-YunxiNeural` |
| Chinese (Trad.) 🇹🇼| HsiaoChen| Female | `zh-TW-HsiaoChenNeural` |
| Chinese (Trad.) 🇹🇼| YunJhe | Male | `zh-TW-YunJheNeural` |

## Technical Flow 🔧
```mermaid
sequenceDiagram
    User->>+Gradio App: Select Language & Voice
    User->>+Gradio App: Input Text
    Gradio App->>+edge-tts: Send Request
    edge-tts-->>-Gradio App: Return Audio
    Gradio App->>+User: Play Audio
````

## Local Development 💻

```bash
# Clone the repository
git clone [https://huggingface.co/spaces/snackshell/selam-tts](https://huggingface.co/spaces/snackshell/selam-tts)
cd selam-tts

# Install dependencies
pip install -r requirements.txt

# Run the app
python app.py
```

## Acknowledgments 🙏

  - Voice Technology: Microsoft Edge TTS
  - Hosting: Hugging Face Spaces
  - UI Framework: Gradio
  - Maintained by: [Bana Codes](https://t.me/banacodes)

[🐛 Report Issues](https://github.com/snackshell/eng-amh-tts/issues)

```
```