An Open Source text-to-speech system built by inverting Whisper.

Llama3.1 learns to Listen

Awesome speech/audio LLMs, representation learning, and codec models

CAIMAN-ASR: low-latency speech recognition on FPGA

🔊 A comprehensive list of open-source datasets for voice and sound computing (95+ datasets).

A playbook for systematically maximizing the performance of deep learning models.

Cost aware hyperparameter tuning algorithm

A bleeding-edge, lock-free, wait-free, continuation-stealing tasking library built on C++20's coroutines

A powerful data & AI notebook templates catalog: prompts, plugins, models, workflow automation, analytics, code snippets - following the IMO framework to be searchable and reusable in any context.

Silero VAD: pre-trained enterprise-grade Voice Activity Detector

Building blocks for foundation models.

A PyTorch implementation of DeepSpeech and DeepSpeech2.

GPTQLoRA: Efficient Finetuning of Quantized LLMs with GPTQ

An efficient OpenFST-based tool for calculating WER and aligning two transcript sequences.

An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

Code for paper: "QuIP: 2-Bit Quantization of Large Language Models With Guarantees"

Foundational Models for State-of-the-Art Speech and Text Translation

4 bits quantization of LLaMA using GPTQ

Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".

CUDA-Warp RNN-Transducer

RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference,…

A curated list of resources dedicated to table recognition

Can we use explanations to improve hate speech models? Our paper accepted at AAAI 2021 tries to explore that question.

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

A curated list of awesome Machine Learning frameworks, libraries and software.

MSc Research project (6 months). Data Assimilation using Deep Learning (AEs). Imperial College Machine Learning MSc 2018-19

Reference implementation of real-time autoregressive wavenet inference

PyTorch CTC Decoder bindings