All 53 articles, sorted alphabetically
Adaptive Jitter Buffer Design
Trade latency for smoothness without overshooting.
Read article →AAC
How AAC (Advanced Audio Coding) provides better quality than MP3 at same bitrate, becoming streaming standard.
Read article →Acoustic echo cancellation architecture
Deep-dive on AEC: loopback reference capture, delay estimation, partitioned-block frequency-domain NLMS, double-talk detection, residual echo suppress…
Read article →Automatic gain control architecture
Deep-dive on automatic gain control (AGC) for voice: level estimation, the gain-computer control loop, fast-attack/slow-release dynamics, the noise ga…
Read article →ASR
How ASR systems convert speech to text, and the modern neural approaches replacing older HMM systems.
Read article →Automatic Speech Recognition Architecture in Depth
A 2500-word walkthrough of ASR architecture: audio input, feature extraction, encoder-decoder, streaming decoding, Whisper family, diarization, VAD, p…
Read article →Microphone beamforming architecture
Deep-dive on microphone beamforming: mic array, delay-and-sum, adaptive MVDR/GSC, DOA estimation, noise, echo, speech enhancement.
Read article →Audio Buffer Management for Low Latency
Underruns, drift correction, and why ring buffers beat queues for audio.
Read article →Clock-drift compensation architecture
Deep-dive on clock-drift compensation in real-time audio: why independent sample clocks inevitably diverge, inferring the parts-per-million drift from…
Read article →Audio Codecs
How audio codecs (MP3, AAC, Opus) compress audio using psychoacoustic models.
Read article →Audio dereverberation architecture
Deep-dive on speech dereverberation: why reverberation is convolution not additive noise, STFT analysis, late-reflection estimation with weighted pred…
Read article →Speaker Diarization
How diarization identifies different speakers in an audio recording (meeting transcription, call analytics).
Read article →Speaker diarization architecture
Deep-dive on speaker diarization: VAD, segments, ECAPA/x-vector embed, spectral clustering, change point, overlap, labels, ASR fusion.
Read article →Speaker diarization v2 architecture
Deep-dive on speaker diarization v2: VAD, segmentation, speaker embeddings, clustering, speaker turn detection, streaming, overlap.
Read article →DTX and comfort noise architecture
Deep-dive on discontinuous transmission and comfort noise generation in voice codecs: a VAD forking speech from silence, DTX stopping full-frame trans…
Read article →Audio forward error correction
Deep-dive on FEC and redundancy for real-time audio: why retransmission fails against the conversational deadline, in-band Opus FEC and RED-style redu…
Read article →Acoustic howling suppression architecture
Deep-dive on howling/feedback suppression: loop-gain stability and maximum/added stable gain, feedback detection and notch filtering, frequency shifti…
Read article →Adaptive jitter buffer architecture
Deep-dive on adaptive jitter buffers: RTP reordering, percentile-based delay estimation, accelerate/expand time-stretching, packet-loss concealment, c…
Read article →Audio Loudness Normalization
LUFS over peak, and why every platform requires it.
Read article →Loudness normalization architecture
Deep-dive on loudness normalization: why peak normalization fails perceptually, the ITU-R BS.1770 chain (K-weighting, 400ms block mean-square, absolut…
Read article →Multi-party audio mixing architecture
Deep-dive on server-side multi-party audio mixing (the MCU): per-stream jitter buffering and decoding, resampling to a common clock to correct device …
Read article →Music generation architecture
Deep-dive on music generation: audio tokenizer, acoustic transformer, vocoder, style conditioning, watermark, rights, streaming.
Read article →Neural audio codecs -- learned compression and audio tokens
Deep-dive on neural audio codecs: learned compression, the encoder-quantizer-decoder architecture, residual vector quantization (RVQ), the bitrate-vs-…
Read article →Real-time noise suppression architecture
Deep-dive on streaming noise suppression: framing and STFT features, causal GRU/TCN mask estimators (RNNoise, DTLN, DeepFilterNet), post-filter smooth…
Read article →Opus codec architecture
Deep-dive on the Opus audio codec: SILK speech and CELT music engines, hybrid mode and content-adaptive switching, flexible frame sizes, FEC and PLC p…
Read article →Packet loss concealment
Deep-dive on packet loss concealment (PLC) for real-time voice: manufacturing replacement frames for lost packets by extrapolating from recent history…
Read article →PCM Audio
How PCM represents analog audio as discrete samples, and how sample rate + bit depth determine quality.
Read article →Real-Time Audio Architecture in Depth
A 2500-word walkthrough of real-time audio: capture, AEC, noise, VAD, Opus codec, packetize + FEC, transport, jitter buffer, decoder, ASR, barge-in.
Read article →Sample-rate conversion architecture
Deep-dive on audio sample-rate conversion (resampling): how to move audio between rates like 48 kHz, 44.1 kHz, and 16 kHz without aliasing or imaging.…
Read article →Audio source separation architecture
Deep-dive on audio source separation: STFT, neural model (U-Net/TasNet), masks, iSTFT, streaming causal, targeting, metrics.
Read article →Spatial audio -- placing sound in 3D space
Deep-dive on spatial audio: the brain's localization cues (ITD, ILD, spectral), the HRTF, binaural rendering, object-based au…
Read article →Audio Streaming
How audio streaming works: adaptive chunk-based (HLS/DASH) and continuous (Icecast, Shoutcast).
Read article →Streaming ASR architecture
Deep-dive on streaming speech recognition: VAD, streaming encoder, CTC/transducer, partials, endpointing, rescore, and observability.
Read article →Audio Streaming Protocols Compared
HLS, DASH, Icecast, WebRTC — pick by use case.
Read article →Audio super-resolution -- reconstructing the missing high frequencies
Deep-dive on audio super-resolution (bandwidth extension): the narrowband problem, adding missing high frequencies, learned reconstruction (not upsamp…
Read article →Text-to-Speech (TTS)
How modern TTS systems generate natural-sounding speech, from concatenative to neural TTS.
Read article →TTS architecture
Deep-dive on text-to-speech: text encoder, acoustic model, vocoder, streaming, voice cloning with consent + watermark, emotion, metrics.
Read article →Voice Activity Detection (VAD)
How VAD identifies when someone is speaking versus silence or background noise.
Read article →Voice Activity Detection architecture
Deep-dive on VAD: capture and framing, energy vs GMM vs neural (Silero) scoring, hysteresis thresholds, hangover and pre-roll smoothing, ASR gating, b…
Read article →Wake word detection architecture
Deep-dive on wake word detection: mic, VAD, KWS model, trigger, false accepts + rejects, user tuning, privacy, multi-language.
Read article →OpenAI Whisper
How Whisper does speech recognition across languages with strong noise robustness.
Read article →Echo Cancellation for Speakerphone
Adaptive filter design with NLMS and frequency-domain implementations.
Read article →Jitter Buffer Design
Smoothing arrival jitter without adding too much latency.
Read article →MP3 vs AAC vs Opus for Streaming
Codec choice for podcasts music and voice.
Read article →The Opus Codec for Real-Time Voice
Why Opus is the default codec for WebRTC and how to tune it for voice vs music.
Read article →Opus Codec Parameters That Matter
Bitrate, complexity, FEC — pick by use case.
Read article →Packet Loss Concealment Algorithms
Reconstructing missing audio packets for VoIP.
Read article →Real-Time Audio Pipelines in Rust
cpal crate ring buffers and avoiding allocations on the audio thread.
Read article →Sample Rate Conversion Techniques
Sinc-based vs FFT-based resampling tradeoffs.
Read article →Speech-to-Text Stack in 2026
Whisper, AssemblyAI, Deepgram — the current trade-offs.
Read article →Voice Activity Detection (VAD)
When to send silence, when to detect speech.
Read article →WebRTC Audio Pipeline End to End
What happens between the mic and the speaker.
Read article →