Supported languages
Palabra API supports the following languages for speech recognition and translation.
- Languages with codes in both columns support both Recognition (as source language) and Translation (as target language).
- Languages marked with
-
in the Recognition column cannot be used as source languages for speech recognition, but translations into these languages are supported. - Languages marked with
-
in the Translation column cannot be used as target languages for translation, but translations from these languages are supported. - Languages marked with
-
in the Voice creation / Voice cloning column cannot be used as a source for voice creation, and auto voice cloning is not available for them.
Auto language detection
To enable automatic language detection, you must manually set asr_model
to alpha
in the transcription
settings.
This is an experimental feature and currently supports only the following languages: .
Please note that ASR quality may be lower than with the default settings for some languages.
"pipeline": {
"transcription": {
"asr_model": "alpha", # <-- required
"source_language": "auto", <--
"detectable_languages": ["en", "fr"], <-- also for this option to work
....
Languages list
Language | Recognition (source) | Translation (target) | Voice creation / Voice cloning |
---|