Skip to content
Thursday, 19 November 2026
Tenesys AI News
Subscribe
Voice AI· Important· 🧪 Worth Testing

AssemblyAI releases STT-3.7 with expanded language support and improved diarization

In short: AssemblyAI has launched STT-3.7, a new speech-to-text model replacing the Universal family. It offers increased language coverage and significant improvements to speaker identification accuracy.

Source: AssemblyAIAssemblyAIAssemblyAI-STT-3.7Original article ↗

This summary was generated automatically by AI from AssemblyAI's publication. It is our own text, not a copy of the original — facts, figures and quotes belong to the source, linked above and below.

What changed?

  • 1Language support increased from 18 to 32 languages.
  • 2New speaker diarization model reduces Diarization Error Rate (DER) by 29%.
  • 3Native support for code-switching.
  • 4Pricing remains unchanged at $0.21 per hour.
AssemblyAI-STT-3.7
ParameterBeforeNow
Supported languages1832
Diarization Error Rate (DER)Not specified29% lower
Price$0.21/hr$0.21/hr

Why it matters

For developers building voice agents and call analytics, the reduced diarization error rate directly correlates to cleaner, more actionable transcriptions when multiple people speak in a single call.

What it means for AI agents and contact centers

When testing this model against current pipelines, prioritize benchmarking the speaker diarization accuracy on overlapping speech, as the 29% improvement in DER could drastically simplify downstream conversation analysis logic.

🧪 Worth Testing

The reduction in diarization errors makes this a key update for improving call analysis quality without increasing processing costs.

AssemblyAI-STT-3.7· New

Sources

  • AssemblyAIOfficialPrimary source
    „AssemblyAI-STT-3.7 is now generally available, with 32 languages and an improved speaker diarization model“
    Original article →
  • AssemblyAIOfficial
    „Which AssemblyAI-STT-3.7 model should you use?“
    Original article →
Published by source
—
Found by our system
16 Nov 2026, 23:08
Summary generated
16 Nov 2026, 23:08

This article was written by AI from the original source. Facts, numbers and prices come from the source; missing values are marked “Not specified”. Legal notice, copyright and privacy

AssemblyAI releases STT-3.7 with expanded language support and improved diarization · TENESYS AI NEWS