transducer

Speech-to-text server framework with next-gen Kaldi

Zero-cost state-machine library for robust, testable and portable user interfaces (most machines compile ~1-2KB)

Rust re-implementation of OpenFST - library for constructing, combining, optimizing, and searching weighted finite-state transducers (FSTs). A Python binding is also available.

A simple debug library for Clojure(Script) that features data-oriented logging and tracing

PyTorch Implementations for End-to-End Automatic Speech Recognition

Memory efficient transducer loss computation

An efficient implementation of RNN-T Prefix Beam Search in C++/CUDA.

Tiny Transducer: A Highly-Efficient Speech Recognition Model on Edge Devices

Streaming speech recognition toolkit: log-mel features, CTC greedy and prefix-beam decoding, transducer-style frame-synchronous emission, forced alignment, Chinese/English CER-WER metrics, and a chunked streaming pipeline with latency accounting. NumPy core with a CPU-only torch extra; trainable demos on synthetic audio, fully offline.

High-performance speech-to-text (ASR) inference and serving engine for NVIDIA GPUs: Whisper, Conformer, Zipformer, RNN-T, Paraformer, Qwen2-Audio. OpenAI-compatible API, streaming, CUDA kernels.

Learn how to write your own transducers. A complement from my blog post serie "Build Your Own Transducer and Impress Your Cat".

This is ultrasonic pest repellent device for bugs based on ATTINY85, resistors and PAM4803 amplifier with set of 40kHz ultrasonic transducers

Target speaker automatic speech recognition (TS-ASR)

🌁 Nondeterministic Finite State Automata for Java (in plain English: flowcharts with multiple possible outcomes)

Calculation tool for loudspeaker design

Code for ICML25 Paper "Overcoming Non-monotonicity in Transducer-based Streaming Generation"

Fast parallel RNN-Transducer.

A PHP transducers implementation in a fluent way

Elixir module that implements a generic finite state transducer with customizable rules expressed in a DSL.

Do lookups on an FST in Python!