Version 12.10

Q1 2026 (Jan–Mar 2026)

Broader TTS Language Support via Automatic Normalization Fallback
Text-to-Speech now supports Italian, Japanese, Brazilian Portuguese, and Mandarin Chinese even in cases where SESTEK's own text-normalization isn't yet available for that language, by automatically falling back to the voice provider's built-in normalization. All previously supported languages and behavior remain unchanged.
Adjustable Speech Tempo After Synthesis
TTS output can now have its speaking speed adjusted after synthesis via a dedicated audio-processing step, producing cleaner results with no added noise or distortion — particularly useful for voices that speak more slowly by nature.
Flexible, Interval-Based Credit Consumption for TTS Licensing
Usage-based TTS licenses can now define a configurable "roundup interval" (for example, every 60 seconds of use equals one credit), giving customers more transparent and flexible consumption-based billing. Licenses without this setting continue to work exactly as before.
Single-Payload Text Submission for Cosy & Kokoro Voices
Cosy and Kokoro voices now support submitting input text as a single request payload, offering a simpler integration option for applicable use cases.

Complete Voice Metadata for Cosy Voices
Cosy voices now return full metadata — including gender and sample rate — through the API, ensuring they display and behave correctly everywhere voices can be selected, with no impact on existing integrations.
Automated, Version-Controlled Database Updates for TTS
Introduced an automated database migration framework for TTS, ensuring schema changes are applied consistently and safely across environments. Existing abbreviation data was migrated to the new framework with no loss of configuration.

Fixed Default Abbreviation Updates Not Applying to Existing Deployments
Resolved an issue where updates to the default abbreviation file were not automatically picked up on deployments that already existed.
Fixed Incorrect Expansion Override Fields After Abbreviation Migration
Resolved an issue where, after abbreviation migration, expansion override fields incorrectly appeared populated for all entries.
Fixed Duplicate Default Abbreviations Appearing as Custom Entries
Resolved an issue where some default abbreviation rows were incorrectly duplicated as custom (user-added) entries during migration.
Fixed a Language Code Error for Kokoro and Cosy Voices
Resolved a language code error affecting Kokoro and Cosy voices.
Fixed an Error Adding Multiple Abbreviations to the Default List at Once
Resolved an error that occurred when multiple abbreviations were added to the default list at the same time.