代码拉取完成,页面将自动刷新
package openai
import (
"bytes"
"context"
"fmt"
"io"
"net/http"
"os"
utils "gitee.com/ledao/go-openai/internal"
)
// Whisper Defines the models provided by OpenAI to use when processing audio with OpenAI.
const (
Whisper1 = "whisper-1"
)
// Response formats; Whisper uses AudioResponseFormatJSON by default.
type AudioResponseFormat string
const (
AudioResponseFormatJSON AudioResponseFormat = "json"
AudioResponseFormatText AudioResponseFormat = "text"
AudioResponseFormatSRT AudioResponseFormat = "srt"
AudioResponseFormatVerboseJSON AudioResponseFormat = "verbose_json"
AudioResponseFormatVTT AudioResponseFormat = "vtt"
)
type TranscriptionTimestampGranularity string
const (
TranscriptionTimestampGranularityWord TranscriptionTimestampGranularity = "word"
TranscriptionTimestampGranularitySegment TranscriptionTimestampGranularity = "segment"
)
// AudioRequest represents a request structure for audio API.
type AudioRequest struct {
Model string
// FilePath is either an existing file in your filesystem or a filename representing the contents of Reader.
FilePath string
// Reader is an optional io.Reader when you do not want to use an existing file.
Reader io.Reader
Prompt string
Temperature float32
Language string // Only for transcription.
Format AudioResponseFormat
TimestampGranularities []TranscriptionTimestampGranularity // Only for transcription.
}
// AudioResponse represents a response structure for audio API.
type AudioResponse struct {
Task string `json:"task"`
Language string `json:"language"`
Duration float64 `json:"duration"`
Segments []struct {
ID int `json:"id"`
Seek int `json:"seek"`
Start float64 `json:"start"`
End float64 `json:"end"`
Text string `json:"text"`
Tokens []int `json:"tokens"`
Temperature float64 `json:"temperature"`
AvgLogprob float64 `json:"avg_logprob"`
CompressionRatio float64 `json:"compression_ratio"`
NoSpeechProb float64 `json:"no_speech_prob"`
Transient bool `json:"transient"`
} `json:"segments"`
Words []struct {
Word string `json:"word"`
Start float64 `json:"start"`
End float64 `json:"end"`
} `json:"words"`
Text string `json:"text"`
httpHeader
}
type audioTextResponse struct {
Text string `json:"text"`
httpHeader
}
func (r *audioTextResponse) ToAudioResponse() AudioResponse {
return AudioResponse{
Text: r.Text,
httpHeader: r.httpHeader,
}
}
// CreateTranscription — API call to create a transcription. Returns transcribed text.
func (c *Client) CreateTranscription(
ctx context.Context,
request AudioRequest,
) (response AudioResponse, err error) {
return c.callAudioAPI(ctx, request, "transcriptions")
}
// CreateTranslation — API call to translate audio into English.
func (c *Client) CreateTranslation(
ctx context.Context,
request AudioRequest,
) (response AudioResponse, err error) {
return c.callAudioAPI(ctx, request, "translations")
}
// callAudioAPI — API call to an audio endpoint.
func (c *Client) callAudioAPI(
ctx context.Context,
request AudioRequest,
endpointSuffix string,
) (response AudioResponse, err error) {
var formBody bytes.Buffer
builder := c.createFormBuilder(&formBody)
if err = audioMultipartForm(request, builder); err != nil {
return AudioResponse{}, err
}
urlSuffix := fmt.Sprintf("/audio/%s", endpointSuffix)
req, err := c.newRequest(
ctx,
http.MethodPost,
c.fullURL(urlSuffix, withModel(request.Model)),
withBody(&formBody),
withContentType(builder.FormDataContentType()),
)
if err != nil {
return AudioResponse{}, err
}
if request.HasJSONResponse() {
err = c.sendRequest(req, &response)
} else {
var textResponse audioTextResponse
err = c.sendRequest(req, &textResponse)
response = textResponse.ToAudioResponse()
}
if err != nil {
return AudioResponse{}, err
}
return
}
// HasJSONResponse returns true if the response format is JSON.
func (r AudioRequest) HasJSONResponse() bool {
return r.Format == "" || r.Format == AudioResponseFormatJSON || r.Format == AudioResponseFormatVerboseJSON
}
// audioMultipartForm creates a form with audio file contents and the name of the model to use for
// audio processing.
func audioMultipartForm(request AudioRequest, b utils.FormBuilder) error {
err := createFileField(request, b)
if err != nil {
return err
}
err = b.WriteField("model", request.Model)
if err != nil {
return fmt.Errorf("writing model name: %w", err)
}
// Create a form field for the prompt (if provided)
if request.Prompt != "" {
err = b.WriteField("prompt", request.Prompt)
if err != nil {
return fmt.Errorf("writing prompt: %w", err)
}
}
// Create a form field for the format (if provided)
if request.Format != "" {
err = b.WriteField("response_format", string(request.Format))
if err != nil {
return fmt.Errorf("writing format: %w", err)
}
}
// Create a form field for the temperature (if provided)
if request.Temperature != 0 {
err = b.WriteField("temperature", fmt.Sprintf("%.2f", request.Temperature))
if err != nil {
return fmt.Errorf("writing temperature: %w", err)
}
}
// Create a form field for the language (if provided)
if request.Language != "" {
err = b.WriteField("language", request.Language)
if err != nil {
return fmt.Errorf("writing language: %w", err)
}
}
if len(request.TimestampGranularities) > 0 {
for _, tg := range request.TimestampGranularities {
err = b.WriteField("timestamp_granularities[]", string(tg))
if err != nil {
return fmt.Errorf("writing timestamp_granularities[]: %w", err)
}
}
}
// Close the multipart writer
return b.Close()
}
// createFileField creates the "file" form field from either an existing file or by using the reader.
func createFileField(request AudioRequest, b utils.FormBuilder) error {
if request.Reader != nil {
err := b.CreateFormFileReader("file", request.Reader, request.FilePath)
if err != nil {
return fmt.Errorf("creating form using reader: %w", err)
}
return nil
}
f, err := os.Open(request.FilePath)
if err != nil {
return fmt.Errorf("opening audio file: %w", err)
}
defer f.Close()
err = b.CreateFormFile("file", f)
if err != nil {
return fmt.Errorf("creating form file: %w", err)
}
return nil
}
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。