feat(audio): count whisper-1 quota by audio duration

common/audio/audio.go

@@ -0,0 +1,19 @@
+package audio
+
+import (
+	"bytes"
+	"context"
+	"os/exec"
+	"strconv"
+)
+
+// GetAudioDuration returns the duration of an audio file in seconds.
+func GetAudioDuration(ctx context.Context, filename string) (float64, error) {
+	// ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 {{input}}
+	c := exec.CommandContext(ctx, "ffprobe", "-v", "error", "-show_entries", "format=duration", "-of", "default=noprint_wrappers=1:nokey=1", filename)
+	output, err := c.Output()
+	if err != nil {
+		return 0, err
+	}
+	return strconv.ParseFloat(string(bytes.TrimSpace(output)), 64)
+}

common/file.go

@@ -0,0 +1,20 @@
+package common
+
+import (
+	"io"
+	"os"
+)
+
+// SaveTmpFile saves data to a temporary file. The filename would be apppended with a random string.
+func SaveTmpFile(filename string, data io.Reader) (string, error) {
+	f, err := os.CreateTemp(os.TempDir(), filename)
+	if err != nil {
+		return "", err
+	}
+	defer f.Close()
+	_, err = io.Copy(f, data)
+	if err != nil {
+		return "", err
+	}
+	return f.Name(), nil
+}

common/http.go

@@ -0,0 +1,22 @@
+package common
+
+import (
+	"bytes"
+	"io"
+	"net/http"
+)
+
+func CloneRequest(old *http.Request) *http.Request {
+	req := old.Clone(old.Context())
+	oldBody, err := io.ReadAll(old.Body)
+	if err != nil {
+		return nil
+	}
+	err = old.Body.Close()
+	if err != nil {
+		return nil
+	}
+	old.Body = io.NopCloser(bytes.NewBuffer(oldBody))
+	req.Body = io.NopCloser(bytes.NewBuffer(oldBody))
+	return req
+}

common/model-ratio.go

@@ -61,7 +61,7 @@ var ModelRatio = map[string]float64{
 	"text-davinci-003":          10,
 	"text-davinci-edit-001":     10,
 	"code-davinci-edit-001":     10,
-	"whisper-1":                 15,  // $0.006 / minute -> $0.006 / 150 words -> $0.006 / 200 tokens -> $0.03 / 1k tokens
+	"whisper-1":                 1,   // $0.006 / minute -> $0.002 / 20 seconds -> $0.002 / 1K tokens
 	"tts-1":                     7.5, // $0.015 / 1K characters
 	"tts-1-1106":                7.5,
 	"tts-1-hd":                  15, // $0.030 / 1K characters
@@ -157,5 +157,8 @@ func GetCompletionRatio(name string) float64 {
 	if strings.HasPrefix(name, "claude-2") {
 		return 2.965517
 	}
+	if strings.HasPrefix(name, "whisper-1") {
+		return 0 // only count input audio duration
+	}
 	return 1
 }