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.
基于 BirdCLEF 2026 的声音鸟类识别课程项目，包含语音特征工程、传统机器学习、CNN/CRNN 对比、GroupKFold 泛化分析与 Gradio 可视化演示。
Audio signal analysis & feature extraction (STFT / log-Mel / MFCC), audio data cleaning pipeline, and MusicGen / Demucs inference benchmarks on a 4GB laptop GPU.
Rust-based Real-time audio DSP frontend for log-Mel feature extraction, streaming, WAV input, stress reports, and its Criterion benchmarks.
To associate your repository with the log-mel topic, visit your repo's landing page and select "manage topics."