Mandarin–Malay–Tamil–English
The four-language pack uses one recognition model for mixed English, Mandarin, Malay and Tamil speech. It covers the four official languages of Singapore.
A fixed multilingual set
The model was trained on conversational audio with language changes. Its documented coverage includes Singaporean English and Singlish, regional Mandarin, Malay and Singapore Tamil. One request returns the mixed transcript in spoken order.Speechmatics: English, Mandarin, Malay and Tamil model
Practical guidance. Use this pack when the conversation can include any of these four languages. It is distinct from the earlier three bilingual models, each of which limits recognition to English and one regional language.Speechmatics: Southeast Asian language model development
Specifications & access
| Detail | Value and conditions |
|---|---|
| Launch date | |
| Access | API, Provider product |
| Model identifiers | cmn_en_ms_taEnglish Mandarin Malay TamilFour-language pack |
| Language selector | cmn_en_ms_taSpeechmatics documentation: Languages and medical coverage |
| Model choices | Standard and Enhanced operating pointsSpeechmatics: English, Mandarin, Malay and Tamil model |
Related models
- Melia 1
Speechmatics · Melia
Multilingual speech recognition model that follows language changes within a recording.
Launch date API · Provider product - Malay–English Bilingual
Speechmatics · Bilingual recognition
Bilingual recognition model for Malay conversation mixed with English terms.
Launch date API · Provider product - Mandarin–English Bilingual
Speechmatics · Bilingual recognition
Bilingual recognition model trained for Mandarin and English code-switching in Southeast Asia.
Launch date API · Provider product - Tamil–English Bilingual
Speechmatics · Bilingual recognition
Bilingual recognition model for Tamil and English speech in the same conversation.
Launch date API · Provider product