Spanish–English Bilingual
The Spanish–English model was available in a December 2023 container release. The February 2024 technical account describes its Ursa foundation and recognition across both languages.
Ursa with a retrained language component
The technical account says the model reused an acoustic component from the monolingual pipeline. The language-model component was retrained on both vocabularies. It did not require a separate language detector to switch between Spanish and English.Speechmatics: Spanish–English model and Ursa ancestry
Practical guidance. This release shows why a bilingual model can be a real trained adaptation even when an API selects it with a domain field. The changed language component affects which mixed-language word sequences the system can recognize.Speechmatics: Spanish–English model and Ursa ancestry
Specifications & access
| Detail | Value and conditions |
|---|---|
| Launch date | |
| Access | API, Provider product |
| Model identifiers | Spanish Englishes / bilingual-en |
| Request settings | language: es; domain: bilingual-enSpeechmatics documentation: Languages and medical coverage |
| Self-hosted access | GPU deployment with the Spanish bilingual client imageSpeechmatics documentation: Accessing container images |
Related models
- Melia 1
Speechmatics · Melia
Multilingual speech recognition model that follows language changes within a recording.
Launch date API · Provider product - Spanish Medical Model
Speechmatics · Enhanced Medical
Spanish clinical recognition model for specialist vocabulary and patient conversations.
Launch date API · Provider product - Ursa
Speechmatics · Ursa
Historical recognition generation with self-supervised audio representations and separate decoding components.
Launch date Access not confirmed - Enhanced
Speechmatics · Enhanced
General speech recognition model that prioritizes accuracy across supported language packs.
Launch date not confirmedAPI · Provider product