go-whisper-api/api/openai.go

package api

import (
    "encoding/json"
    "net/http"
    "strings"
    "time"
)

func (s *Server) handleOpenAITranscriptions(w http.ResponseWriter, r *http.Request) {
    if r.Method != http.MethodPost {
        writeOpenAIError(w, http.StatusMethodNotAllowed, "method not allowed")
        return
    }
    if err := r.ParseMultipartForm(128 << 20); err != nil {
        writeOpenAIError(w, http.StatusBadRequest, err.Error())
        return
    }
    modelID, err := s.models.Resolve(r.FormValue("model"), s.cfg.DefaultModel)
    if err != nil {
        writeOpenAIError(w, http.StatusBadRequest, err.Error())
        return
    }
    modelPath, err := s.models.Path(modelID)
    if err != nil {
        writeOpenAIError(w, http.StatusBadRequest, err.Error())
        return
    }
    audioPath, cleanup, err := s.saveUploadedOpenAI(r)
    if err != nil {
        writeOpenAIError(w, http.StatusBadRequest, err.Error())
        return
    }
    defer cleanup()

    stt := s.parseOpenAISTTOptions(r)
    result, err := s.transcribe(r.Context(), modelPath, audioPath, stt)
    if err != nil {
        writeOpenAIError(w, http.StatusInternalServerError, err.Error())
        return
    }

    switch strings.ToLower(strings.TrimSpace(r.FormValue("response_format"))) {
    case "text":
        w.Header().Set("Content-Type", "text/plain; charset=utf-8")
        w.WriteHeader(http.StatusOK)
        _, _ = w.Write([]byte(result.Text))
    default:
        writeJSON(w, http.StatusOK, map[string]string{"text": result.Text})
    }
}

func (s *Server) handleOpenAIModels(w http.ResponseWriter, r *http.Request) {
    if r.Method != http.MethodGet {
        writeOpenAIError(w, http.StatusMethodNotAllowed, "method not allowed")
        return
    }
    ids, err := s.models.List()
    if err != nil {
        writeOpenAIError(w, http.StatusInternalServerError, err.Error())
        return
    }
    now := time.Now().Unix()
    data := make([]map[string]any, 0, len(ids))
    for _, id := range ids {
        data = append(data, map[string]any{
            "id":       id,
            "object":   "model",
            "created":  now,
            "owned_by": "go-whisper-api",
        })
    }
    writeJSON(w, http.StatusOK, map[string]any{
        "object": "list",
        "data":   data,
    })
}

func (s *Server) parseOpenAISTTOptions(r *http.Request) sttOptions {
    lang := strings.TrimSpace(r.FormValue("language"))
    if lang == "" {
        lang = s.cfg.Language
    }
    return sttOptions{
        language:  lang,
        punctuate: s.punctCfg.ShouldApplyAPI(r, s.cfg.DefaultPunctuation),
    }
}

func (s *Server) saveUploadedOpenAI(r *http.Request) (path string, cleanup func(), err error) {
    if r.MultipartForm == nil {
        if err := r.ParseMultipartForm(128 << 20); err != nil {
            return "", nil, err
        }
    }
    return saveUploadedRawFields(r, []string{"file", "audio", "wav"})
}

func writeOpenAIError(w http.ResponseWriter, code int, msg string) {
    w.Header().Set("Content-Type", "application/json; charset=utf-8")
    w.WriteHeader(code)
    _ = json.NewEncoder(w).Encode(map[string]any{
        "error": map[string]any{
            "message": msg,
            "type":    openAIErrorType(code),
        },
    })
}

func openAIErrorType(code int) string {
    switch code {
    case http.StatusBadRequest:
        return "invalid_request_error"
    case http.StatusUnauthorized:
        return "authentication_error"
    case http.StatusNotFound:
        return "not_found_error"
    default:
        return "server_error"
    }
}