2026.09.30 Batch SaaS
Fixes
Fix for English transcription with the Enhanced model: spoken numbers like "ten thousand" were sometimes incorrectly transcribed as "ten zero zero zero".
Updated Orchestrator Version: 2026.09.25+62a2863062+16.2.0
Release Notes
Follow new updates and improvements to Speechmatics.
2026.09.23 Batch SaaS
Improvements
Improved accuracy and latency for Dutch language with the Enhanced model when processing non-speech audio.
The default behaviour for polling
GET /jobs/{jobid}andGET /jobs/{jobid}/transcriptendpoints is changing so that requests can wait for up to 5 seconds for a job to complete before returning. This behaviour can be disabled by sendingwait=0to return immediately. Client-side polling can be avoided for most jobs by using Synchronous transcription.
Updated Orchestrator Version: 2026.09.21+8a4b0316ed+16.1.0
2026.09.22 Batch SaaS
Improvements
Enhanced and Standard models - significantly faster processing of audio files up to 20 minutes in duration for the following languages: Arabic/English, Bashkir, Basque, Belarusian, Bengali, Bulgarian, Czech, English/Malay, English/Tamil, Esperanto, Estonian, Galician, Interlingua, Irish, Lithuanian, Maltese, Mandarin/English, Mandarin/English/Malay/Tamil, Marathi, Mongolian, Romanian, Slovakian, Slovenian, Swahili, Tagalog, Tamil, Thai, Turkish, Urdu, Uyghur and Welsh
Earlier updates
- 2026.09.09 Batch SaaSImprovements Improved accuracy for GPU Enhanced English when transcribing recent years, e.g., “2026”. Updated Orchestrator Version: 2026.09.07+1f5eba77f2+16.0.0Batch SaaS
- 2026.09.03 Batch SaaSImprovements Updated Melia 1 model for multilingual transcription provides: 2% relative WER improvement across many languages Substantially fewer transliteration errors on Hindi-English audio Better handling of Hindi-English code-switching within a single sentence More accurate transcription of consecutive alphanumerics, such as repeated letters and digits Updated Orchestrator Version:2026.07.31+f3542ffa97+1.5.0Batch SaaS
- 2026.08.07 Batch SaaSImprovements The default behaviour for polling GET /jobs/{jobid} and GET /jobs/{jobid}/transcript endpoints is changing so that requests can wait for up to 2 seconds for a job to complete before returning. This reduces the need for frequent client-side polling and can improve turnaround time for users. This behaviour can be disabled by sending wait=0 to return immediately. Client-side polling can be avoided for most jobs by using Synchronous transcription.Batch SaaS
- 2026.08.06 Batch SaaSFixes Fixed recently introduced issue with remove_disfluencies, where punctuation around disfluencies was not being removed. Disfluency tagging (and removal) is currently unavailable for Finnish (fi), Norwegian (no) and Swedish (sv), after being added on Jul 28, 2027. Bug introduced in release version: 2026.07.28 Updated Orchestrator Version: 2026.08.05+3c8ca0561c+15.18.0Batch SaaS
- 2026.08.03 Batch SaaSImprovements Updated Melia 1 model for multilingual transcription provides: Improved transcription accuracy (WER reduced by 2% relative overall) Improved alphanumerics, with output now more closely matching written form Updated Orchestrator Version: 2026.07.31+f3542ffa97+1.5.0Batch SaaS
- 2026.07.28 Batch SaaSNew Disfluency detection and removal now supports 19 additional languages, including Arabic, Danish, Dutch, Finnish, French, German, Greek, Hebrew, Hindi, Hungarian, Italian, Japanese, Mandarin, Norwegian, Polish, Portuguese, Russian, Spanish, and Swedish. See the documentation for more details. Updated Orchestrator Version: 2026.07.24+b3c1e3bbc5+15.17.0Batch SaaS
- 2026.07.14 Batch SaaSImprovements New Melia 1 model for multilingual transcription provides: Improved English transcription accuracy (WER reduced by 2.4% relative) Improved code-switching accuracy with language hints Improved alphanumeric formatting Updated Orchestrator Version: 2026.07.09+8723c03fc9+1.4.0Batch SaaS