155 lines
5.5 KiB
Markdown
155 lines
5.5 KiB
Markdown
|
---
|
|||
|
|
base_model: canopylabs/3b-hi-ft-research_release
|
|||
|
|
license: apache-2.0
|
|||
|
|
language:
|
|||
|
|
- hi # Hindi
|
|||
|
|
- bn # Bengali
|
|||
|
|
- mr # Marathi
|
|||
|
|
- te # Telugu
|
|||
|
|
- kn # Kannada
|
|||
|
|
- bho # Bhojpuri
|
|||
|
|
- mag # Magahi
|
|||
|
|
- hne # Chhattisgarhi
|
|||
|
|
- mai # Maithili
|
|||
|
|
- as # Assamese
|
|||
|
|
- brx # Bodo
|
|||
|
|
- doi # Dogri
|
|||
|
|
- gu # Gujarati
|
|||
|
|
- ml # Malayalam
|
|||
|
|
- pa # Punjabi
|
|||
|
|
- ta # Tamil
|
|||
|
|
- ne # Nepali
|
|||
|
|
- sa # Sanskrit
|
|||
|
|
- en # English (Indian)
|
|||
|
|
tags:
|
|||
|
|
- text-to-speech
|
|||
|
|
- speech-synthesis
|
|||
|
|
- transformers
|
|||
|
|
- multilingual
|
|||
|
|
- indic
|
|||
|
|
- orpheus
|
|||
|
|
- lora
|
|||
|
|
- low-latency
|
|||
|
|
- gguf
|
|||
|
|
- zero-shot
|
|||
|
|
- emotions
|
|||
|
|
- discrete-audio-tokens
|
|||
|
|
task_categories:
|
|||
|
|
- text-to-speech
|
|||
|
|
pipeline_tag: text-to-speech
|
|||
|
|
pretty_name: Svara-TTS v1
|
|||
|
|
datasets:
|
|||
|
|
- SYSPIN
|
|||
|
|
- RASA
|
|||
|
|
- IndicTTS
|
|||
|
|
- SPICOR
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
# svara-TTS v1 — Open Multilingual TTS for India’s Voices
|
|||
|
|
|
|||
|
|
[](https://huggingface.co/kenpath/svara-tts-v1)
|
|||
|
|
[](https://huggingface.co/spaces/kenpath/svara-tts)
|
|||
|
|
[](https://colab.research.google.com/drive/15YxFo1DzdQNbFUIZ1HJA4AN4oHqKxGtg)
|
|||
|
|
[](https://github.com/Kenpath/svara-tts-inference)
|
|||
|
|
|
|||
|
|
**svara-TTS** is a developer-first multilingual TTS model for **19 languages** (18 Indic + Indian English).
|
|||
|
|
Built on an Orpheus-style discrete audio token approach, it targets **clarity, expressiveness, and low-latency** on commodity GPUs/CPUs.
|
|||
|
|
It supports light-weight **emotion/style control** (e.g., `<happy>`, `<sad>`, `<anger>`, `<fear>`) and simple **speaker identities** (`Language (Gender)`), with **zero-shot** adaptation paths.
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## At a Glance
|
|||
|
|
|
|||
|
|
- **Languages (19):** Hindi, Bengali, Marathi, Telugu, Kannada, Bhojpuri, Magahi, Chhattisgarhi, Maithili, Assamese, Bodo, Dogri, Gujarati, Malayalam, Punjabi, Tamil, Nepali, Sanskrit, Indian English.
|
|||
|
|
- **Expressivity:** End-of-utterance style tags; natural prosody; code-switch aware.
|
|||
|
|
- **Latency & Deployment:** Works well with **GGUF** exports; suitable for edge/CPU scenarios.
|
|||
|
|
- **Adaptability:** **LoRA-friendly** for quick speaker/domain specialization.
|
|||
|
|
|
|||
|
|
Try it live on the [**Demo Space**](https://huggingface.co/spaces/kenpath/svara-tts), or on **[Colab](https://colab.research.google.com/drive/15YxFo1DzdQNbFUIZ1HJA4AN4oHqKxGtg)**
|
|||
|
|
Deployment scripts and **inference repo** will be available soon. Watch our [Github](https://github.com/Kenpath/svara-tts-inference) for updates
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## Prompting (Orpheus-style)
|
|||
|
|
|
|||
|
|
- Place style/emotion tags **at the end** of the sentence:
|
|||
|
|
`आज... सच में अच्छी खबर है — शाम को मिलते हैं! <happy>`
|
|||
|
|
- Use punctuation to hint prosody (ellipses, commas, exclamation).
|
|||
|
|
- For technical or dense text, end with `<clear>` to prioritize intelligibility.
|
|||
|
|
|
|||
|
|
> Speaker IDs follow a simple convention: **`Language (Gender)`** (e.g., `Marathi (Male)`).
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## Training Data Summary
|
|||
|
|
|
|||
|
|
Trained on **2000+ hours** of open, high-quality speech from **SYSPIN**, **RASA**, **IndicTTS**, and **SPICOR**, covering **~50 speakers** (balanced male/female) across **19 languages**.
|
|||
|
|
Data was curated to encourage natural prosody, broad coverage, and stable multilingual transfer. See **Acknowledgments** for provenance.
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## Intended Uses
|
|||
|
|
|
|||
|
|
- Multilingual assistants, IVR, learning apps, reading aids, accessibility tools
|
|||
|
|
- Content localization (education, public-information, civic services)
|
|||
|
|
- Research on Indic prosody, emotion control, cross-lingual transfer
|
|||
|
|
|
|||
|
|
## Out-of-Scope / Not Intended
|
|||
|
|
|
|||
|
|
- Impersonation of private individuals or public figures without consent
|
|||
|
|
- Deceptive content (fraud, harassment, misinformation)
|
|||
|
|
- Safety-critical deployments without human oversight
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## Limitations
|
|||
|
|
|
|||
|
|
- **Proper nouns & rare entities:** may require spelling hints or `<clear>`.
|
|||
|
|
- **Very long sentences:** chunk or add punctuation for natural prosody.
|
|||
|
|
- **Emotion strength:** varies by language due to data density.
|
|||
|
|
- **Code-mixing:** common patterns work; it’s not a deterministic rules engine.
|
|||
|
|
|
|||
|
|
Many of these improve with targeted LoRA finetuning and better preprocessing.
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## Responsible Use
|
|||
|
|
|
|||
|
|
By using this model, you agree to follow applicable laws and ethical guidelines.
|
|||
|
|
Avoid impersonation, harassment, targeted deception, or other harmful uses.
|
|||
|
|
Where appropriate, disclose synthetic speech to end users.
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## Sources & Links
|
|||
|
|
|
|||
|
|
- **Model:** https://huggingface.co/kenpath/svara-tts-v1
|
|||
|
|
- **Demo Space:** https://huggingface.co/spaces/kenpath/svara-tts
|
|||
|
|
- **Inference repo:** https://github.com/Kenpath/svara-tts-inference
|
|||
|
|
- **Colab:** https://colab.research.google.com/drive/15YxFo1DzdQNbFUIZ1HJA4AN4oHqKxGtg
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## 🙏 Acknowledgments
|
|||
|
|
|
|||
|
|
This work was developed by **[Kenpath Technologies](https://kenpath.ai/)** for the open-source community. We also thank RunPod for the startup credits that supported our GPU compute.
|
|||
|
|
|
|||
|
|
- **Canopy Labs — Orpheus:** foundational ideas & open release
|
|||
|
|
Release: https://canopylabs.ai/releases/orpheus_can_speak_any_language
|
|||
|
|
- **SPIRE Lab, IISc Bangalore** — **SYSPIN** (multilingual studio) and **SPICOR** (Indian English)
|
|||
|
|
- **AI4Bharat** — **RASA** expressive speech
|
|||
|
|
- **IIT Madras** — **IndicTTS**
|
|||
|
|
- **Unsloth** — helpful notes & tooling
|
|||
|
|
- **RunPod** — startup GPU credits that accelerated experiments
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## License
|
|||
|
|
|
|||
|
|
**Apache-2.0**
|
|||
|
|
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
## Versioning & Changelog
|
|||
|
|
|
|||
|
|
- **v1.0.0:** Initial public release (19 languages)
|