--- license: cc-by-sa-4.0 datasets: - speechcolab/gigaspeech - parler-tts/mls_eng_10k - reach-vb/jenny_tts_dataset - MikhailT/hifi-tts - ylacombe/expresso - keithito/lj_speech - collabora/ai4bharat-shrutilipi language: - en - hi base_model: - openai-community/gpt2-medium pipeline_tag: text-to-speech library_name: transformers --- | Platform | Link | |----------|------| | 🌎 Live Demo | [indrivoice.ai](https://indrivoice.ai/) | | 𝕏 Twitter | [@11mlabs](https://x.com/11mlabs) | | 🐱 GitHub | [Indri Repository](https://github.com/cmeraki/indri) | | 🤗 Hugging Face (Collection) | [Indri collection](https://huggingface.co/collections/11mlabs/indri-673dd4210b4369037c736bfe) | | 📝 Release Blog | [Release Blog](https://www.indrivoice.ai/blog/2024-11-21-building-indri-tts) | # Model Card for indri-0.1-350m-tts Indri is a series of audio models that can do TTS, ASR, and audio continuation. This is the medium sized model (350M) in our series and supports TTS tasks in 2 languages: 1. English 2. Hindi ## Model Details ### Model Description `indri-0.1-350m-tts` is a novel, small, and lightweight TTS model based on the transformer architecture. It models audio as tokens and can generate high-quality audio with consistent style cloning of the speaker. ### Samples | Text | Sample | | --- | --- | |अतीत गौरवशाली, वर्तमान आशावादी, भविष्य उज्जवल|

| |भाइयों और बहनों, ये हमारा सौभाग्य है कि हम सब मिलकर इस महान देश को नई ऊंचाइयों पर ले जाने का सपना देख रहे हैं।|

| |Hello दोस्तों, future of speech technology mein अपका स्वागत है |

| |Artificial Intelligence's collaborative hub: Transforming Machine Learning together|

| |Intelligent machines processing data at lightning-fast electronic speeds|

| ### Key features 1. Extremely efficient, based on GPT-2 medium architecture. The methodology can be extended to any autoregressive transformer-based architecture. 2. Ultra-fast. Using our [self hosted service option](#self-hosted-service), the model can achieve speeds up to 400 toks/s (4s of audio generation per s) and under 20ms time to first token on RTX6000Ada NVIDIA GPU. 1. On RTX6000Ada, it can support a batch size of ~500 with full context length of 1024 tokens 3. Supports voice cloning with small prompts (<5s). 4. Code mixing text input in 2 languages - English and Hindi. ### Details 1. Model Type: GPT-2 based language model 2. Size: 350M parameters 3. Language Support: English, Hindi 4. License: CC BY 4.0 ## Technical details Here's a brief of how the model works: 1. Converts input text into tokens 2. Runs autoregressive decoding on GPT-2 based transformer model and generates audio tokens 3. Decodes audio tokens (using [Kyutai/mimi](https://huggingface.co/kyutai/mimi)) to audio Please read our blog [here](https://www.indrivoice.ai/blog/2024-11-21-building-indri-tts) for more technical details on how it was built. ## How to Get Started with the Model ### 🤗 pipelines Use the code below to get started with the model. Pipelines are the best way to get started with the model. ```python import torch import torchaudio from transformers import pipeline model_id = '11mlabs/indri-0.1-350m-tts' task = 'indri-tts' pipe = pipeline( task, model=model_id, device=torch.device('cuda:0'), # Update this based on your hardware, trust_remote_code=True ) output = pipe(['Hi, my name is Indri and I like to talk.']) torchaudio.save('output.wav', output[0]['audio'][0], sample_rate=24000) ``` ### Self hosted service ```bash git clone https://github.com/cmeraki/indri.git cd indri pip install -r requirements.txt # Install ffmpeg (for Mac/Windows, refer here: https://www.ffmpeg.org/download.html) sudo apt update -y sudo apt upgrade -y sudo apt install ffmpeg -y python -m inference --model_path 11mlabs/indri-0.1-350m-tts --device cuda:0 --port 8000 ``` ## Citation If you use this model in your research, please cite: ```bibtex @misc{indri-multimodal-alm, author = {11mlabs}, title = {Indri: Multimodal audio language model}, year = {2024}, publisher = {GitHub}, journal = {GitHub Repository}, howpublished = {\url{https://github.com/cmeraki/indri}}, email = {compute@merakilabs.com} } ``` ## BibTex 1. [nanoGPT](https://github.com/karpathy/nanoGPT) 2. [Kyutai/mimi](https://huggingface.co/kyutai/mimi) ```bibtex @techreport{kyutai2024moshi, title={Moshi: a speech-text foundation model for real-time dialogue}, author={Alexandre D\'efossez and Laurent Mazar\'e and Manu Orsini and Am\'elie Royer and Patrick P\'erez and Herv\'e J\'egou and Edouard Grave and Neil Zeghidour}, year={2024}, eprint={2410.00037}, archivePrefix={arXiv}, primaryClass={eess.AS}, url={https://arxiv.org/abs/2410.00037}, } ``` 3. [Whisper](https://github.com/openai/whisper) ```bibtex @misc{radford2022whisper, doi = {10.48550/ARXIV.2212.04356}, url = {https://arxiv.org/abs/2212.04356}, author = {Radford, Alec and Kim, Jong Wook and Xu, Tao and Brockman, Greg and McLeavey, Christine and Sutskever, Ilya}, title = {Robust Speech Recognition via Large-Scale Weak Supervision}, publisher = {arXiv}, year = {2022}, copyright = {arXiv.org perpetual, non-exclusive license} } ``` 4. [silero-vad](https://github.com/snakers4/silero-vad) ```bibtex @misc{Silero VAD, author = {Silero Team}, title = {Silero VAD: pre-trained enterprise-grade Voice Activity Detector (VAD), Number Detector and Language Classifier}, year = {2024}, publisher = {GitHub}, journal = {GitHub repository}, howpublished = {\url{https://github.com/snakers4/silero-vad}}, commit = {insert_some_commit_here}, email = {hello@silero.ai} } ```