1// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
2
3package openai
4
5import (
6 "bytes"
7 "context"
8 "io"
9 "mime/multipart"
10 "net/http"
11
12 "github.com/openai/openai-go/internal/apiform"
13 "github.com/openai/openai-go/internal/apijson"
14 "github.com/openai/openai-go/internal/requestconfig"
15 "github.com/openai/openai-go/option"
16 "github.com/openai/openai-go/packages/param"
17 "github.com/openai/openai-go/packages/respjson"
18)
19
20// AudioTranslationService contains methods and other services that help with
21// interacting with the openai API.
22//
23// Note, unlike clients, this service does not read variables from the environment
24// automatically. You should not instantiate this service directly, and instead use
25// the [NewAudioTranslationService] method instead.
26type AudioTranslationService struct {
27 Options []option.RequestOption
28}
29
30// NewAudioTranslationService generates a new service that applies the given
31// options to each request. These options are applied after the parent client's
32// options (if there is one), and before any request-specific options.
33func NewAudioTranslationService(opts ...option.RequestOption) (r AudioTranslationService) {
34 r = AudioTranslationService{}
35 r.Options = opts
36 return
37}
38
39// Translates audio into English.
40func (r *AudioTranslationService) New(ctx context.Context, body AudioTranslationNewParams, opts ...option.RequestOption) (res *Translation, err error) {
41 opts = append(r.Options[:], opts...)
42 path := "audio/translations"
43 err = requestconfig.ExecuteNewRequest(ctx, http.MethodPost, path, body, &res, opts...)
44 return
45}
46
47type Translation struct {
48 Text string `json:"text,required"`
49 // JSON contains metadata for fields, check presence with [respjson.Field.Valid].
50 JSON struct {
51 Text respjson.Field
52 ExtraFields map[string]respjson.Field
53 raw string
54 } `json:"-"`
55}
56
57// Returns the unmodified JSON received from the API
58func (r Translation) RawJSON() string { return r.JSON.raw }
59func (r *Translation) UnmarshalJSON(data []byte) error {
60 return apijson.UnmarshalRoot(data, r)
61}
62
63type AudioTranslationNewParams struct {
64 // The audio file object (not file name) translate, in one of these formats: flac,
65 // mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
66 File io.Reader `json:"file,omitzero,required" format:"binary"`
67 // ID of the model to use. Only `whisper-1` (which is powered by our open source
68 // Whisper V2 model) is currently available.
69 Model AudioModel `json:"model,omitzero,required"`
70 // An optional text to guide the model's style or continue a previous audio
71 // segment. The
72 // [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
73 // should be in English.
74 Prompt param.Opt[string] `json:"prompt,omitzero"`
75 // The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
76 // output more random, while lower values like 0.2 will make it more focused and
77 // deterministic. If set to 0, the model will use
78 // [log probability](https://en.wikipedia.org/wiki/Log_probability) to
79 // automatically increase the temperature until certain thresholds are hit.
80 Temperature param.Opt[float64] `json:"temperature,omitzero"`
81 // The format of the output, in one of these options: `json`, `text`, `srt`,
82 // `verbose_json`, or `vtt`.
83 //
84 // Any of "json", "text", "srt", "verbose_json", "vtt".
85 ResponseFormat AudioTranslationNewParamsResponseFormat `json:"response_format,omitzero"`
86 paramObj
87}
88
89func (r AudioTranslationNewParams) MarshalMultipart() (data []byte, contentType string, err error) {
90 buf := bytes.NewBuffer(nil)
91 writer := multipart.NewWriter(buf)
92 err = apiform.MarshalRoot(r, writer)
93 if err == nil {
94 err = apiform.WriteExtras(writer, r.ExtraFields())
95 }
96 if err != nil {
97 writer.Close()
98 return nil, "", err
99 }
100 err = writer.Close()
101 if err != nil {
102 return nil, "", err
103 }
104 return buf.Bytes(), writer.FormDataContentType(), nil
105}
106
107// The format of the output, in one of these options: `json`, `text`, `srt`,
108// `verbose_json`, or `vtt`.
109type AudioTranslationNewParamsResponseFormat string
110
111const (
112 AudioTranslationNewParamsResponseFormatJSON AudioTranslationNewParamsResponseFormat = "json"
113 AudioTranslationNewParamsResponseFormatText AudioTranslationNewParamsResponseFormat = "text"
114 AudioTranslationNewParamsResponseFormatSRT AudioTranslationNewParamsResponseFormat = "srt"
115 AudioTranslationNewParamsResponseFormatVerboseJSON AudioTranslationNewParamsResponseFormat = "verbose_json"
116 AudioTranslationNewParamsResponseFormatVTT AudioTranslationNewParamsResponseFormat = "vtt"
117)