← Skill graph
Speech recognition
Part of Speech interfaces and voice agents
Needs first
Helps first
Nothing recommended beyond what is required.
Unlocks
Where it is assessed
10 assessed items, 0 done.
- AIE-106 M1.7 Three systems, one voice · tier 1
- AIE-106 M1.8 Retention: ASR architectures and their assumptions · tier 1
- AIE-106 M2.4 Word error rate is not enough · tier 1
- AIE-106 M2.5 The errors that matter · tier 2
- AIE-106 M2.6 Retention: recognition metrics and their blind spots · tier 1
- AIE-106 M3.4 Build the eval set nobody has · tier 1
- AIE-106 M3.5 Benchmark the vendors on it · tier 1
- AIE-106 M3.6 Retention: accent, code-switching and evaluation coverage · tier 1
- AIE-106 M8.2 The latency and accuracy report · tier 2
- AIE-106 M8.4 Oral defense · tier 4
Verified sources
- CS244S @ Stanford: LLM Based Spoken Language Processing 2025 · Gridspace
- Welcome to the Hugging Face Audio course!
- Evaluation metrics for ASR
- On the End-to-End Solution to Mandarin-English Code-switching Speech Recognition
- Common Voice: A Massively-Multilingual Speech Corpus
- Conformer: Convolution-augmented Transformer for Speech Recognition
- CS-Dialogue: A 104-Hour Dataset of Spontaneous Mandarin-English Code-Switching Dialogues for Speech Recognition
- HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units
- SeamlessM4T: Massively Multilingual & Multimodal Machine Translation
- wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations
- Robust Speech Recognition via Large-Scale Weak Supervision
- WhisperX: Time-Accurate Speech Transcription of Long-Form Audio
- openai/whisper
- Speech and Language Processing