← Back
Home › Systems › AI applications

speech-recognition

speech recognition

Whisper

SystemTechniques · 4Primitives · 28 speech-recognition RNN transducer (RNN-T)machine-learning · medium: Streaming sequence model that sums all monotone alignments in a frame-by-label lattice and conditions each label on earlier ones.›Beam search decodingml-training · medium: Keep the k best partial sequences by summed log-probability at every step, set finished ones aside, and return the best (n-best) by a length-normalised score.›Dynamic batchingml-training · medium: Queue independent requests and dispatch them as one batch when it is full or the oldest has waited max_delay; bounded extra latency buys throughput.›Sample rate conversionsignal-processing · medium: Polyphase and windowed-sinc resampling; asynchronous rate conversion for clock driftShort-time Fourier transform and sp…signal-processing · low: Slide a window over a signal and FFT each frame to get a time-frequency picture; the front end of audio ML, vibration and radar analysis.›Byte-pair encoding (BPE) tokenizati…machine-learning · low: Learn a subword vocabulary by repeatedly merging the most frequent adjacent pair; byte-level BPE encodes any input losslessly with 32k-200k tokens.›Connectionist temporal classificati…machine-learning · low: Alignment-free sequence loss that sums the probability of every frame-to-label alignment with a blank symbol, using a dynamic program.›Data augmentationmachine-learning · low: Generate label-preserving variants of each training example on the fly to enlarge the effective dataset and bake in invariances.›Ngram language modelmachine-learning · low: N-gram counts with Kneser-Ney smoothing, compact storage (KenLM-style) and perplexityText normalizationmachine-learning · low: Rewrite messy text into one canonical form for matchers and models, or into spoken words for speech output (and back), keeping a map to the original offsets.›Voice activity detectionmachine-learning · low: Per-frame speech/non-speech decision with smoothing, from an energy threshold over a tracked noise floor up to small neural nets.›Convolutional networksml-training · low: Layers that slide a small shared kernel over a grid, giving translation equivariance and few parameters; still the most compute-efficient choice for many vision tasks.›Transformer blockml-training · low: The decoder-only pre-norm block (attention plus MLP on a residual stream) behind every modern LLM, with exact shapes, parameter, FLOP and KV-cache counts.›Weight quantizationml-training · low: Store weights as 8/4-bit integers with per-group scales; decode is memory-bound, so 4-bit weights give about 3-4x tokens/s and fit 4x larger models.›Fast Fourier transformnumerics · low: Computes the DFT in O(n log n) by recursively splitting into even and odd halves; the basis of fast convolution and spectral methods.›Edit distancesearch-sort · low: Levenshtein/Damerau distance by DP, bit-parallel Myers variantFinite state transducersearch-sort · low: Minimal acyclic FST mapping sorted keys to values; compact term dictionaries with prefix/fuzzy searchRecurrent networksmachine-learning · low: GRU and LSTM cells for streaming sequence modelsd-ary heapdata-structures · low: Array heap with d children per node (d = 4 typical): half the binary heap's depth, siblings in one cache line, faster priority queues.›Top-k selectionsearch-sort · low: Keep the k best of n items without sorting all n - bounded min-heap for streams, quickselect for arrays, threshold pruning to skip most items.›+ 12 more primitives

Similar systems

Share the most building blocks with this one.

auto-captions 32 %prompt-injection-detection 19 %mobile-keyboard-ime 19 %realtime-audio-dsp 19 %machine-translation 18 %