Create README.md

README.md

@@ -0,0 +1,100 @@
+---
+license: apache-2.0
+language: tr
+tags:
+- text-to-speech
+- turkish
+- tts
+- f5-tts
+datasets:
+- custom
+pipeline_tag: text-to-speech
+---
+
+# Orkhon-TTS
+
+## Model Details
+
+*   **Model Type:** Text-to-Speech (TTS)
+*   **Architecture:** F5 TTS
+*   **Language:** Turkish (tr)
+*   **Developed by:** Hasan Can Solakoğlu
+*   **Model Version:** v1.0 (Alpha)
+*   **License:** Apache License 2.0
+*   **Demo:** [Orkhon-TTS Hugging Face Space](https://huggingface.co/spaces/hcsolakoglu/Orkhon-TTS)
+
+## Model Description
+
+Orkhon-TTS is a Turkish Text-to-Speech model based on the F5 TTS architecture. It has been trained by Hasan Can Solakoğlu. This model is currently in its alpha stage.
+
+The primary goal of this model is to provide a high-quality Turkish TTS voice for researchers, companies, and students.
+
+### Voice Cloning Capabilities
+
+This model possesses voice cloning capabilities. Users are expected to use these features responsibly and ethically.
+
+## Training Data
+
+Model was trained on high-quality single-speaker Turkish speech data. All training data was meticulously prepared and curated by Hasan Can Solakoğlu.
+
+
+## Intended Uses & Limitations
+
+### Intended Uses
+
+*   Generating Turkish speech from text for various applications.
+*   Research in Turkish speech synthesis.
+*   Educational purposes for understanding TTS models.
+*   Prototyping voice-enabled applications for Turkish users.
+
+### Limitations and Bias
+
+*   **Alpha Stage:** The model is currently in an alpha stage, meaning it may produce artifacts or unnatural-sounding speech in some cases.
+*   **Pronunciation of Abbreviations and Numbers:** The current version may not optimally handle the pronunciation of abbreviations or the verbalization of numbers written in digit form. These are planned improvements for v2.
+*   **Single Speaker:** The current public model is based on a single speaker.
+
+### Out-of-Scope Uses
+
+*   Generating speech for illegal or unethical purposes.
+*   Impersonating individuals without their explicit consent.
+*   Creating hate speech or misleading content.
+
+## How to Get Started with the Model
+
+Usage instructions and code examples will be provided in the repository associated with this model. For a live demo, please visit the [Hugging Face Space](https://huggingface.co/spaces/hcsolakoglu/Orkhon-TTS).
+
+## Future Plans (v2)
+
+*   Improved handling of abbreviations.
+*   Enhanced pronunciation of numbers (reading numerical digits).
+*   Training on a larger dataset.
+*   Longer training duration for potentially higher quality.
+
+## Author Contact
+
+For questions or feedback about the model, please contact Hasan Can Solakoğlu via [Twitter/X](https://x.com/HCSolakoglu).
+
+## Citation
+
+If you use this model in your research or project, please consider citing it (details to be provided upon official release or publication).
+
+```bibtex
+@misc{orkhon_tts_hcsolakoglu_2025,
+    author = {Solakoğlu, Hasan Can},
+    title = {Orkhon-TTS: A Turkish Text-to-Speech Model},
+    year = {2025},
+    publisher = {Hugging Face},
+    journal = {Hugging Face Model Hub},
+    howpublished = {\url{https://huggingface.co/hcsolakoglu/Orkhon-TTS}} 
+}
+```
+
+## Disclaimer
+
+This model is provided "as-is" without any warranty, express or implied. The developers and contributors are not responsible for any damages or losses arising from the use of this model.
+
+**Responsible Use of Voice Cloning:** The voice cloning capabilities of this model must be used responsibly and ethically. Users are solely responsible for the content they generate and any consequences arising from its use. The model creator (Hasan Can Solakoğlu) cannot be held liable for any misuse of the model or its outputs. Do not use this model to impersonate individuals without their explicit consent or for any malicious purposes.
+
+## License
+
+This model is licensed under the Apache License, Version 2.0. See the [LICENSE](LICENSE) file for more details.