nullAdd translation to empty string values and empty "keyWords" arrays. The source value…
- User
nullAdd translation to empty string values and empty "keyWords" arrays. The source value to the translation is already translated (non-empty) values of neighboring properties. The language of individual translations is determined by the key of the translated value. Delete source (non-empty) localization properties (for example "en-US"), for size reduction of the result.
The following rules apply to the "keyWords" section (only if it exists in input JSON):
- Translations contain a array of synonymous terms in string format. These strings enter as the first parameter fo JavaScript function RegExp() (see en-US example).
- The number of synonyms varies according to the language of the translation.
- The basic idea of synonymous terms is described by the name of the key (endOfDictationDefault, acceptance, rejection, ...), but do not forget to add synonyms for individual terms in sourse languages arrays.
- Some RegExp fields contain all expressions ending with the suffix " _+$".
- Letters with diacritics must also contain an alternative without diacritics (eg: /[áa]/i).
The following rules apply to the "wakeWords" section (only if it exists in input JSON):
- They are English names for which there is no translation, so a phonetic translation is necessary.
Result must by valid JSON. Do not add or modify keys in the JSON structure. Display only updated JSON string without your comments. Update this input JSON string translations: ``` { "wakeWords": { "_jeyjei": { "sk-SK": "", "en-US": "Jeyjei" }, "_jo_oo_jee": { "sk-SK": "", "en-US": "Jooojee" }, "_jin_jehre": { "sk-SK": "", "en-US": "Jinjehre" }, "_jujeene": { "sk-SK": "", "en-US": "Jujeene" }, "_jaju": { "sk-SK": "", "en-US": "Jaju" }, "alexa": { "sk-SK": "", "en-US": "Alexa" }, "hey jarvis": { "sk-SK": "", "en-US": "Hey jarvis" }, "hey rhasspy": { "sk-SK": "", "en-US": "Hey rhasspy" } } }
- Assistant
{ "wakeWords": { "_jeyjei": { "sk-SK": "Džejdžei" }, "_jo_oo_jee": { "sk-SK": "Džóódží" }, "_jin_jehre": { "sk-SK": "Džindžere" }, "_jujeene": { "sk-SK": "Džudžín" }, "_jaju": { "sk-SK": "Džajdžu" }, "alexa": { "sk-SK": "Alexa" }, "hey jarvis": { "sk-SK": "Hej Džarvis" }, "hey rhasspy": { "sk-SK": "Hej Raspí" } } }
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.