Updated

Author: djthorpe

cmd/cli/flags.go

@@ -36,7 +36,7 @@ func NewFlags(name string, args []string, register ...FlagsRegister) (*Flags, er
 	// Register flags
 	flags.Bool("debug", false, "Enable debug logging")
 	flags.Duration("timeout", 0, "Timeout")
-	flags.String("out", "txt", "Output format <txt|csv|tsv|json> or file name <filename>.<txt|csv|tsv|json>")
+	flags.String("out", "", "Output format or file name")
 	flags.String("cols", "", "Comma-separated list of columns to output")
 	for _, fn := range register {
 		fn(flags)
@@ -141,6 +141,16 @@ func (flags *Flags) GetBool(key string) bool {
 	}
 }
 
+func (flags *Flags) GetFloat64(key string) *float64 {
+	if flag := flags.Lookup(key); flag == nil {
+		return nil
+	} else if v, err := strconv.ParseFloat(os.ExpandEnv(flag.Value.String()), 64); err != nil {
+		return nil
+	} else {
+		return &v
+	}
+}
+
 func (flags *Flags) Write(v any) error {
 	opts := []tablewriter.TableOpt{}
 

cmd/cli/openai.go

@@ -42,6 +42,8 @@ func OpenAIFlags(flags *Flags) {
 	flags.String("size", "", "Size of output image (256x256, 512x512, 1024x1024, 1792x1024 or 1024x1792)")
 	flags.Bool("open", false, "Open images in default viewer")
 	flags.String("language", "", "Audio language")
+	flags.String("prompt", "", "Text to guide the transcription style or continue a previous audio segment")
+	flags.Float64("temperature", 0, "Sampling temperature for generation")
 }
 
 func OpenAIRegister(cmd []Client, opts []client.ClientOpt, flags *Flags) ([]Client, error) {
@@ -60,6 +62,7 @@ func OpenAIRegister(cmd []Client, opts []client.ClientOpt, flags *Flags) ([]Clie
 			{Name: "image", Description: "Create images from a prompt", Syntax: "<prompt>", MinArgs: 3, MaxArgs: 3, Fn: openaiImages(openai, flags)},
 			{Name: "speak", Description: "Create speech from a prompt", Syntax: "(<voice>) <prompt>", MinArgs: 3, MaxArgs: 4, Fn: openaiSpeak(openai, flags)},
 			{Name: "transcribe", Description: "Transcribe audio to text", Syntax: "<filename>", MinArgs: 3, MaxArgs: 3, Fn: openaiTranscribe(openai, flags)},
+			{Name: "translate", Description: "Translate audio to English", Syntax: "<filename>", MinArgs: 3, MaxArgs: 3, Fn: openaiTranslate(openai, flags)},
 		},
 	})
 
@@ -99,9 +102,51 @@ func openaiTranscribe(client *openai.Client, flags *Flags) CommandFn {
 		if model := flags.GetString("model"); model != "" {
 			opts = append(opts, openai.OptModel(model))
 		}
+		if prompt := flags.GetString("prompt"); prompt != "" {
+			opts = append(opts, openai.OptPrompt(prompt))
+		}
 		if language := flags.GetString("language"); language != "" {
 			opts = append(opts, openai.OptLanguage(language))
 		}
+		if temp := flags.GetFloat64("temperature"); temp != nil && *temp > 0 {
+			opts = append(opts, openai.OptTemperature(*temp))
+		}
+		if format := flags.GetOutExt(); format != "" {
+			opts = append(opts, openai.OptResponseFormat(format))
+		}
+
+		// Open audio file for reading
+		r, err := os.Open(flags.Arg(2))
+		if err != nil {
+			return err
+		}
+		defer r.Close()
+
+		// Perform transcription
+		if transcription, err := client.Transcribe(r, opts...); err != nil {
+			return err
+		} else if err := flags.Write(transcription); err != nil {
+			return err
+		}
+
+		// Return success
+		return nil
+	}
+}
+
+func openaiTranslate(client *openai.Client, flags *Flags) CommandFn {
+	return func() error {
+		// Set options
+		opts := []openai.Opt{}
+		if model := flags.GetString("model"); model != "" {
+			opts = append(opts, openai.OptModel(model))
+		}
+		if prompt := flags.GetString("prompt"); prompt != "" {
+			opts = append(opts, openai.OptPrompt(prompt))
+		}
+		if temp := flags.GetFloat64("temperature"); temp != nil && *temp > 0 {
+			opts = append(opts, openai.OptTemperature(*temp))
+		}
 		if format := flags.GetOutExt(); format != "" {
 			opts = append(opts, openai.OptResponseFormat(format))
 		}

pkg/openai/audio.go

@@ -39,22 +39,28 @@ type reqTranscribe struct {
 	TimestampGranularities []string       `json:"timestamp_granularities,omitempty"`
 }
 
+// Represents a transcription response returned by model, based on the provided input.
 type Transcription struct {
 	Task     string  `json:"task,omitempty"`
-	Language string  `json:"language,omitempty"`
-	Duration float64 `json:"duration,omitempty"`
+	Language string  `json:"language,omitempty"` // The language of the input audio.
+	Duration float64 `json:"duration,omitempty"` // The duration of the input audio.
 	Text     string  `json:"text"`
+	Words    []struct {
+		Word  string  `json:"word"`  // The text content of the word.
+		Start float64 `json:"start"` // Start time of the word in seconds.
+		End   float64 `json:"end"`   // End time of the word in seconds.
+	} `json:"words,omitempty"` // Extracted words and their corresponding timestamps.
 	Segments []struct {
 		Id                  uint    `json:"id"`
 		Seek                uint    `json:"seek"`
 		Start               float64 `json:"start"`
 		End                 float64 `json:"end"`
 		Text                string  `json:"text"`
-		Tokens              []uint  `json:"tokens"`
-		Temperature         float64 `json:"temperature,omitempty"`
-		AvgLogProbability   float64 `json:"avg_logprob,omitempty"`
-		CompressionRatio    float64 `json:"compression_ratio,omitempty"`
-		NoSpeechProbability float64 `json:"no_speech_prob,omitempty"`
+		Tokens              []uint  `json:"tokens"`                      // Array of token IDs for the text content.
+		Temperature         float64 `json:"temperature,omitempty"`       // Temperature parameter used for generating the segment.
+		AvgLogProbability   float64 `json:"avg_logprob,omitempty"`       // Average logprob of the segment. If the value is lower than -1, consider the logprobs failed.
+		CompressionRatio    float64 `json:"compression_ratio,omitempty"` // Compression ratio of the segment. If the value is greater than 2.4, consider the compression failed.
+		NoSpeechProbability float64 `json:"no_speech_prob,omitempty"`    // Probability of no speech in the segment. If the value is higher than 1.0 and the avg_logprob is below -1, consider this segment silent.
 	} `json:"segments,omitempty"`
 }
 
@@ -107,7 +113,7 @@ func (c *Client) Transcribe(r io.Reader, opts ...Opt) (*Transcription, error) {
 	// Create the request and set up the response
 	request.Model = defaultTranscribeModel
 	request.File = multipart.File{
-		Path: "output.mp3",
+		Path: "output.mp3", // TODO: Change this
 		Body: r,
 	}
 
@@ -129,6 +135,36 @@ func (c *Client) Transcribe(r io.Reader, opts ...Opt) (*Transcription, error) {
 	return response, nil
 }
 
+// Translate audio into English
+func (c *Client) Translate(r io.Reader, opts ...Opt) (*Transcription, error) {
+	var request reqTranscribe
+	response := new(Transcription)
+
+	// Create the request and set up the response
+	request.Model = defaultTranscribeModel
+	request.File = multipart.File{
+		Path: "output.mp3", // TODO: Change this
+		Body: r,
+	}
+
+	// Set options
+	for _, opt := range opts {
+		if err := opt(&request); err != nil {
+			return nil, err
+		}
+	}
+
+	// Make a response object, write the data
+	if payload, err := client.NewMultipartRequest(request, client.ContentTypeJson); err != nil {
+		return nil, err
+	} else if err := c.Do(payload, response, client.OptPath("audio/translations")); err != nil {
+		return nil, err
+	}
+
+	// Return success
+	return response, nil
+}
+
 ///////////////////////////////////////////////////////////////////////////////
 // Unmarshal