* Discord & Telegram support ASR through groq

pkg/channels/discord.go

@@ -3,17 +3,26 @@ package channels
 import (
 	"context"
 	"fmt"
+	"io"
+	"log"
+	"net/http"
+	"os"
+	"path/filepath"
+	"strings"
+	"time"
 
 	"github.com/bwmarrin/discordgo"
 	"github.com/sipeed/picoclaw/pkg/bus"
 	"github.com/sipeed/picoclaw/pkg/config"
 	"github.com/sipeed/picoclaw/pkg/logger"
+	"github.com/sipeed/picoclaw/pkg/voice"
 )
 
 type DiscordChannel struct {
 	*BaseChannel
-	session *discordgo.Session
-	config  config.DiscordConfig
+	session     *discordgo.Session
+	config      config.DiscordConfig
+	transcriber *voice.GroqTranscriber
 }
 
 func NewDiscordChannel(cfg config.DiscordConfig, bus *bus.MessageBus) (*DiscordChannel, error) {
@@ -28,9 +37,14 @@ func NewDiscordChannel(cfg config.DiscordConfig, bus *bus.MessageBus) (*DiscordC
 		BaseChannel: base,
 		session:     session,
 		config:      cfg,
+		transcriber: nil,
 	}, nil
 }
 
+func (c *DiscordChannel) SetTranscriber(transcriber *voice.GroqTranscriber) {
+	c.transcriber = transcriber
+}
+
 func (c *DiscordChannel) Start(ctx context.Context) error {
 	logger.InfoC("discord", "Starting Discord bot")
 
@@ -103,11 +117,48 @@ func (c *DiscordChannel) handleMessage(s *discordgo.Session, m *discordgo.Messag
 	mediaPaths := []string{}
 
 	for _, attachment := range m.Attachments {
-		mediaPaths = append(mediaPaths, attachment.URL)
-		if content != "" {
-			content += "\n"
+		isAudio := isAudioFile(attachment.Filename, attachment.ContentType)
+
+		if isAudio {
+			localPath := c.downloadAttachment(attachment.URL, attachment.Filename)
+			if localPath != "" {
+				mediaPaths = append(mediaPaths, localPath)
+
+				transcribedText := ""
+				if c.transcriber != nil && c.transcriber.IsAvailable() {
+					ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
+					defer cancel()
+
+					result, err := c.transcriber.Transcribe(ctx, localPath)
+					if err != nil {
+						log.Printf("Voice transcription failed: %v", err)
+						transcribedText = fmt.Sprintf("[audio: %s (transcription failed)]", localPath)
+					} else {
+						transcribedText = fmt.Sprintf("[audio transcription: %s]", result.Text)
+						log.Printf("Audio transcribed successfully: %s", result.Text)
+					}
+				} else {
+					transcribedText = fmt.Sprintf("[audio: %s]", localPath)
+				}
+
+				if content != "" {
+					content += "\n"
+				}
+				content += transcribedText
+			} else {
+				mediaPaths = append(mediaPaths, attachment.URL)
+				if content != "" {
+					content += "\n"
+				}
+				content += fmt.Sprintf("[attachment: %s]", attachment.URL)
+			}
+		} else {
+			mediaPaths = append(mediaPaths, attachment.URL)
+			if content != "" {
+				content += "\n"
+			}
+			content += fmt.Sprintf("[attachment: %s]", attachment.URL)
 		}
-		content += fmt.Sprintf("[attachment: %s]", attachment.URL)
 	}
 
 	if content == "" && len(mediaPaths) == 0 {
@@ -136,3 +187,60 @@ func (c *DiscordChannel) handleMessage(s *discordgo.Session, m *discordgo.Messag
 
 	c.HandleMessage(senderID, m.ChannelID, content, mediaPaths, metadata)
 }
+
+func isAudioFile(filename, contentType string) bool {
+	audioExtensions := []string{".mp3", ".wav", ".ogg", ".m4a", ".flac", ".aac", ".wma"}
+	audioTypes := []string{"audio/", "application/ogg", "application/x-ogg"}
+
+	for _, ext := range audioExtensions {
+		if strings.HasSuffix(strings.ToLower(filename), ext) {
+			return true
+		}
+	}
+
+	for _, audioType := range audioTypes {
+		if strings.HasPrefix(strings.ToLower(contentType), audioType) {
+			return true
+		}
+	}
+
+	return false
+}
+
+func (c *DiscordChannel) downloadAttachment(url, filename string) string {
+	mediaDir := filepath.Join(os.TempDir(), "picoclaw_media")
+	if err := os.MkdirAll(mediaDir, 0755); err != nil {
+		log.Printf("Failed to create media directory: %v", err)
+		return ""
+	}
+
+	localPath := filepath.Join(mediaDir, filename)
+
+	resp, err := http.Get(url)
+	if err != nil {
+		log.Printf("Failed to download attachment: %v", err)
+		return ""
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		log.Printf("Failed to download attachment, status: %d", resp.StatusCode)
+		return ""
+	}
+
+	out, err := os.Create(localPath)
+	if err != nil {
+		log.Printf("Failed to create file: %v", err)
+		return ""
+	}
+	defer out.Close()
+
+	_, err = io.Copy(out, resp.Body)
+	if err != nil {
+		log.Printf("Failed to write file: %v", err)
+		return ""
+	}
+
+	log.Printf("Attachment downloaded successfully to: %s", localPath)
+	return localPath
+}