the Verli blog

Notes on speech, translation, and building with audio

Hands-on write-ups from building live transcription and translation: benchmarks, tradeoffs, and the data behind the decisions.

A speaker addresses three seated note-takers who are wildly out of sync: one page has run long and spilled to the floor, one is short and crumpled, one has barely begun.
ResearchJuly 24, 2026 · 10 min read

Streaming Speech-to-Text Benchmark: Latency and the Accuracy Cost of Going Live

Ten speech-to-text engines run through their real-time streaming APIs on identical audio: streaming WER against batch WER on matched clips with paired-bootstrap intervals, plus first word and finalization latency.

Illustration of a person speaking while a single coral thread runs to four seated figures, each weaving it into a different pattern on a handloom frame.
ResearchJuly 19, 2026 · 12 min read

Speech-to-Text Accuracy by Language: 16 Engines

Compare 16 speech-to-text engines across 10 languages using the same audio, word error rate, character error rate, and paired-bootstrap tests.

A hand-drawn tailor fits an unfinished teal coat to a customer on warm dotted paper.
GuideJuly 18, 2026 · 16 min read

Text-to-Speech APIs Compared: Pricing and Features

Compare 17 text-to-speech APIs on pricing, streaming, voice cloning, languages, rate limits, training-data policies, latency, and quality.

Three hand-drawn runners move from firm ground into deep sand as their positions change.
ResearchJuly 16, 2026 · 13 min read

Speech-to-Text Accuracy Benchmark: Engine Results

See major speech-to-text engines tested on identical LibriSpeech and Earnings-22 audio with one scorer and paired-bootstrap confidence intervals.

Hand-drawn figures adjust a traveler's backpack before a rough path on warm dotted paper.
GuideJuly 15, 2026 · 14 min read

Choosing a Speech-to-Text API in 2026: Pricing and Features Compared

A data-first look at 17 speech-to-text APIs plus the open-source models you can self-host: per-hour pricing, real-time streaming, built-in translation, languages, diarization, rate limits, deployment, and what open models actually cost to run. No winner named, just the tradeoffs, with sources.

Two hand-drawn people at a cafe table move an extra coral chair out of their conversation.
GuideJuly 15, 2026 · 8 min read

The 6 best real-time meeting translation tools in 2026, compared

A hands-on comparison of six real-time meeting translation tools (Verli, JotMe, Maestra, Transync, Whisperr, Notta) plus what Zoom, Teams, and Google Meet translate natively. Bot versus no-bot, captions versus spoken, and what each costs.