Extract phoneme-level timestamps from speeh audio.
pytorch model for contexless-phoneme prediction from speech audio
A multilingual phoneme recognizer capable of generalizing zero-shot to unseen phoneme inventories.
Speech Assessment API in FastAPI with HuggingFace 🤗
End-to-end IPA-based phoneme recognition pipeline using Wav2Vec2, featuring preprocessing, vocabulary generation, model training, and evaluation.
Self-hosted Docker speech API for OpenAI-compatible transcription and text-to-speech, streaming ASR, voice cloning, phoneme recognition, and MCP.
My bachelor thesis on Phoneme recognition and alignment on the TIMIT dataset
Natural Language Processing Projects
Official Python SDK for the Vocametrix voice analysis API — AVQI, DSI, jitter/shimmer, pronunciation assessment, speech-to-text, prosody similarity, and 40+ more clinical and acoustic measures.
Audio AI system for Quran recitation analysis using phoneme recognition, boundary detection, duration estimation, CTC, Wav2Vec2, and Conformer.
Speech and speaker recognition — MFCC feature extraction, HMM alignment and concatenation, and phoneme recognition. (KTH DT2119)
AI-powered English pronunciation coaching: phoneme- and pitch-level speech analysis with interactive visual feedback. Vue 3 frontend, serverless AWS backend, and Wav2Vec2/SPICE ML model serving (PyTorch/TensorFlow, SageMaker/TorchServe).
End-to-end Urdu Child Speech Sound Disorder (SSD) assessment pipeline achieving 9.47% Test PER via fine-tuned IndicWav2Vec-Hindi CTC phoneme recognition, canonical Urdu G2P conversion, and logit-based Goodness-of-Pronunciation (GOP) alignment.
音素级发音评测引擎。本地运行，毫秒响应，看见每一个声音。核心能力是将一段语音拆解到音素级别，告诉你哪个音读对了、哪个读错了、错在哪里、怎么改。
A wizard game to practice English pronunciation offline. Defeat monsters by pronouncing spell words correctly. Powered by Godot & wav2vec2.
2-People-Spanish-Average-Tone-Speech-Synthesis-Corpus
Context-aware neural decoding for speech BCI: extending DCoND with skip-diphone auxiliary supervision and temporal smoothness regularization on the Brain-to-Text '24 Benchmark.
Record yourself speaking English, see which sounds you got wrong, and get coached on how to fix them.
AI pronunciation coach for English and Mandarin
To associate your repository with the phoneme-recognition topic, visit your repo's landing page and select "manage topics."