TensorSpeech/TensorFlowASR

:zap: TensorFlowASR: Almost State-of-the-art Automatic Speech Recognition in Tensorflow 2. Supported languages that can use characters or subwords

What it solves

It provides a framework for building and training Automatic Speech Recognition (ASR) systems, enabling the conversion of spoken audio into text using various state-of-the-art neural network architectures.

How it works

The project implements several ASR architectures categorized by their training loss functions: Transducer models (using RNNT Loss) like Conformer and ContextNet, and CTC models (using CTC Loss) like DeepSpeech2 and Jasper. It includes tools for feature extraction and data augmentation, and supports converting trained models to TFLite format to optimize memory and computation for deployment.

Who it’s for

Developers and researchers working on speech-to-text applications who prefer the TensorFlow ecosystem and need models that can be deployed to edge devices via TFLite.

Highlights

  • Supports multiple architectures: Conformer, ContextNet, RNN Transducer, DeepSpeech2, and Jasper.
  • Includes streaming capabilities via the Streaming Conformer.
  • Provides a path to TFLite conversion for efficient deployment.
  • Supports diverse datasets, including English (LibriSpeech, Common Voice) and Vietnamese corpora.

Related

  • Project
    lenML/Speech-AI-Forge

    A unified framework and API server providing a centralized interface for multiple TTS and ASR models, including ChatTTS, CosyVoice and Whisper.

  • Project
    tensorflow/neural-structured-learning

    A TensorFlow framework for training neural networks using structured signals, such as graphs and adversarial perturbations, to improve accuracy with limited labeled data and increase robustness.

  • Project
    fishaudio/fish-diffusion

    A diffusion-based training framework for Text-to-Speech, Singing Voice Synthesis, and Singing Voice Conversion that supports multi-speaker training and efficient hardware utilization.

  • Project
    githubharald/SimpleHTR

    A TensorFlow-based handwritten text recognition system that converts images of single words or text lines into digital text using CNN and LSTM layers.

  • Project
    QwenAudio/SenseVoice

    A multilingual speech foundation model that provides high-accuracy ASR, emotion recognition, and audio event detection with exceptionally low inference latency.