You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
110 lines
4.0 KiB
110 lines
4.0 KiB
package audomodel
|
|
|
|
type (
|
|
ISys interface {
|
|
CreateTask(userId string, url string, enableSpeakerInfo bool, language, callbackURL, callbackData string) (taskID string, logID string, err error)
|
|
QueryTask(taskID, xTtLogid string) (statusCode string, contexts []*ContextStruct, err error)
|
|
RecognizeFlash(userId string, url string, enableSpeakerInfo bool, language string) (statusCode string, logID string, contexts []*ContextStruct, err error)
|
|
}
|
|
|
|
// UserStruct represents the user information in the request
|
|
// 用户信息结构体
|
|
UserStruct struct {
|
|
UID string `json:"uid"`
|
|
}
|
|
|
|
// AudioStruct represents the audio file information
|
|
// 音频信息结构体
|
|
AudioStruct struct {
|
|
URL string `json:"url"`
|
|
Language string `json:"language"`
|
|
Format string `json:"format"`
|
|
}
|
|
|
|
// CorpusStruct represents corpus related settings
|
|
// 语料库相关设置结构体
|
|
CorpusStruct struct {
|
|
CorrectTableName string `json:"correct_table_name"`
|
|
Context string `json:"context"`
|
|
}
|
|
|
|
// RequestSettingsStruct represents the specific request parameters
|
|
// 请求参数设置结构体
|
|
RequestSettingsStruct struct {
|
|
ModelName string `json:"model_name"`
|
|
ModelVersion string `json:"model_version"`
|
|
EnableChannelSplit bool `json:"enable_channel_split"`
|
|
EnableDDC bool `json:"enable_ddc"`
|
|
EnableSpeakerInfo bool `json:"enable_speaker_info"`
|
|
EnablePunc bool `json:"enable_punc"`
|
|
EnableITN bool `json:"enable_itn"`
|
|
Corpus CorpusStruct `json:"corpus"`
|
|
}
|
|
|
|
// SubmitRequestStruct represents the full payload for the submit task
|
|
// 提交任务的完整请求结构体
|
|
SubmitRequestStruct struct {
|
|
User UserStruct `json:"user"`
|
|
Audio AudioStruct `json:"audio"`
|
|
Request RequestSettingsStruct `json:"request"`
|
|
Callback string `json:"callback,omitempty"`
|
|
CallbackData string `json:"callback_data,omitempty"`
|
|
}
|
|
// ContextStruct represents a minimal transcript segment
|
|
// 识别结果最小结构：文本片段与可选的时间与说话人
|
|
ContextStruct struct {
|
|
Content string `json:"content"`
|
|
StartTime int64 `json:"start_time,omitempty"`
|
|
EndTime int64 `json:"end_time,omitempty"`
|
|
Speaker string `json:"speaker,omitempty"`
|
|
}
|
|
|
|
// FlashResponseStruct 极速版返回的数据结构
|
|
FlashResponseStruct struct {
|
|
AudioInfo struct {
|
|
Duration int64 `json:"duration"`
|
|
} `json:"audio_info"`
|
|
Result struct {
|
|
Additions struct {
|
|
Duration string `json:"duration"`
|
|
} `json:"additions"`
|
|
Text string `json:"text"`
|
|
Utterances []struct {
|
|
EndTime int64 `json:"end_time"`
|
|
StartTime int64 `json:"start_time"`
|
|
Text string `json:"text"`
|
|
Words []struct {
|
|
Confidence float64 `json:"confidence"`
|
|
EndTime int64 `json:"end_time"`
|
|
StartTime int64 `json:"start_time"`
|
|
Text string `json:"text"`
|
|
} `json:"words"`
|
|
Additions struct {
|
|
ChannelID string `json:"channel_id"`
|
|
Speaker string `json:"speaker"`
|
|
} `json:"additions,omitempty"`
|
|
} `json:"utterances"`
|
|
} `json:"result"`
|
|
}
|
|
)
|
|
|
|
var defsys ISys
|
|
|
|
func OnInit(config map[string]interface{}, option ...Option) (err error) {
|
|
defsys, err = newSys(newOptions(config, option...))
|
|
return
|
|
}
|
|
|
|
func NewSys(option ...Option) (sys ISys, err error) {
|
|
sys, err = newSys(newOptionsByOption(option...))
|
|
return
|
|
}
|
|
func CreateTask(userId string, url string, enableSpeakerInfo bool, language, callbackURL, callbackData string) (taskID string, logID string, err error) {
|
|
return defsys.CreateTask(userId, url, enableSpeakerInfo, language, callbackURL, callbackData)
|
|
}
|
|
func QueryTask(taskID, xTtLogid string) (statusCode string, contexts []*ContextStruct, err error) {
|
|
return defsys.QueryTask(taskID, xTtLogid)
|
|
}
|
|
func RecognizeFlash(userId string, url string, enableSpeakerInfo bool, language string) (statusCode string, logID string, contexts []*ContextStruct, err error) {
|
|
return defsys.RecognizeFlash(userId, url, enableSpeakerInfo, language)
|
|
}
|
|
|