Speech Revolutions vs AssemblyAI

AssemblyAI is batch-first and developer-focused, with audio-intelligence features layered on top — summarisation, topic detection, content moderation. We build transcription, diarization and timestamps only, and that focus is where the measured gap appears.

Where we differ measurably is speaker diarization — telling apart who said what — and that gap is large enough to decide the question for meetings, interviews and contact-centre audio.

Diarization error rate

Lower is better. Overlap-aware at a 0.25 s collar, on public datasets, measured 2026-07-12 against AssemblyAI (their API default at the time of the run).

DatasetSpeech RevolutionsAssemblyAI
AMI-SDM9.7%25.3%
NotSoFar10.9%23.3%
DiPCo26.7%28.9%

The harness that produced these is public, and so are the datasets — so this is a number you can reproduce or contest rather than take on trust. Full benchmarks and methodology.

Differences that matter in practice

Speech RevolutionsAssemblyAI
Audio intelligenceNot offered. Transcription, diarization and timestamps only.Summarisation, topic detection, content moderation.
Progress reportingLive progress over SSE while a file is processed.Poll for status.
Language handlingAlways auto-detects, including switching mid-file.language_code, or language_detection to auto-detect.
Keyword biasingcustom_vocabulary.keyterms_prompt (word_boost is deprecated).

Pricing

We charge $0.003 per audio minute — $0.18 an hour — with speaker diarization, word timestamps, punctuation and keyword biasing included rather than priced separately.

We deliberately do not quote AssemblyAI's prices here. They are tiered, they change without notice, and a stale number on a page like this is worse than no number at all. The comparison worth making is the effective cost of one audio hour with the features you actually need switched on — check theirs against ours directly.

What switching involves

Auth moves from the Authorization header to X-API-Key, and the submit-then-poll loop becomes upload, complete, then either poll or watch the SSE stream. The SDKs expose an AssemblyAI-shaped result.text, so much of a response parser carries over unchanged.

The migration guide maps every field, side by side, in four languages.