Merge branch 'main' of code.byted.org:langmanus/deer-flow

src/prompts/coordinator.md

@@ -2,12 +2,12 @@
 CURRENT_TIME: {{ CURRENT_TIME }}
 ---
 
-You are Deer, a friendly AI assistant. You specialize in handling greetings and small talk, while handing off research tasks to a specialized planner.
+You are DeerFlow, a friendly AI assistant. You specialize in handling greetings and small talk, while handing off research tasks to a specialized planner.
 
 # Details
 
 Your primary responsibilities are:
-- Introducing yourself as Deer when appropriate
+- Introducing yourself as DeerFlow when appropriate
 - Responding to greetings (e.g., "hello", "hi", "good morning")
 - Engaging in small talk (e.g., how are you)
 - Politely rejecting inappropriate or harmful requests (e.g., prompt leaking, harmful content generation)
@@ -47,7 +47,7 @@ Your primary responsibilities are:
 
 # Notes
 
-- Always identify yourself as Deer when relevant
+- Always identify yourself as DeerFlow when relevant
 - Keep responses friendly but professional
 - Don't attempt to solve complex problems or create research plans yourself
 - Maintain the same language as the user

src/server/app.py

@@ -1,19 +1,22 @@
 # Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
 # SPDX-License-Identifier: MIT
 
+import base64
 import json
 import logging
+import os
 from typing import List, cast
 from uuid import uuid4
 
-from fastapi import FastAPI
+from fastapi import FastAPI, HTTPException
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import StreamingResponse
+from fastapi.responses import StreamingResponse, Response
 from langchain_core.messages import AIMessageChunk, ToolMessage
 from langgraph.types import Command
 
 from src.graph.builder import build_graph
-from src.server.chat_request import ChatMessage, ChatRequest
+from src.server.chat_request import ChatMessage, ChatRequest, TTSRequest
+from src.tools import VolcengineTTS
 
 logger = logging.getLogger(__name__)
 
@@ -137,3 +140,59 @@ def _make_event(event_type: str, data: dict[str, any]):
     if data.get("content") == "":
         data.pop("content")
     return f"event: {event_type}\ndata: {json.dumps(data, ensure_ascii=False)}\n\n"
+
+
+@app.post("/api/tts")
+async def text_to_speech(request: TTSRequest):
+    """Convert text to speech using volcengine TTS API."""
+    try:
+        app_id = os.getenv("VOLCENGINE_TTS_APPID", "")
+        if not app_id:
+            raise HTTPException(
+                status_code=400, detail="VOLCENGINE_TTS_APPID is not set"
+            )
+        access_token = os.getenv("VOLCENGINE_TTS_ACCESS_TOKEN", "")
+        if not access_token:
+            raise HTTPException(
+                status_code=400, detail="VOLCENGINE_TTS_ACCESS_TOKEN is not set"
+            )
+        cluster = os.getenv("VOLCENGINE_TTS_CLUSTER", "volcano_tts")
+        voice_type = os.getenv("VOLCENGINE_TTS_VOICE_TYPE", "BV700_V2_streaming")
+
+        tts_client = VolcengineTTS(
+            appid=app_id,
+            access_token=access_token,
+            cluster=cluster,
+            voice_type=voice_type,
+        )
+        # Call the TTS API
+        result = tts_client.text_to_speech(
+            text=request.text[:1024],
+            encoding=request.encoding,
+            speed_ratio=request.speed_ratio,
+            volume_ratio=request.volume_ratio,
+            pitch_ratio=request.pitch_ratio,
+            text_type=request.text_type,
+            with_frontend=request.with_frontend,
+            frontend_type=request.frontend_type,
+        )
+
+        if not result["success"]:
+            raise HTTPException(status_code=500, detail=str(result["error"]))
+
+        # Decode the base64 audio data
+        audio_data = base64.b64decode(result["audio_data"])
+
+        # Return the audio file
+        return Response(
+            content=audio_data,
+            media_type=f"audio/{request.encoding}",
+            headers={
+                "Content-Disposition": (
+                    f"attachment; filename=tts_output.{request.encoding}"
+                )
+            },
+        )
+    except Exception as e:
+        logger.exception(f"Error in TTS endpoint: {str(e)}")
+        raise HTTPException(status_code=500, detail=str(e))

src/server/chat_request.py

@@ -1,7 +1,7 @@
 # Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
 # SPDX-License-Identifier: MIT
 
-from typing import List, Optional, Union
+from typing import List, Optional, Union, Dict, Any
 
 from pydantic import BaseModel, Field
 
@@ -44,3 +44,19 @@ class ChatRequest(BaseModel):
     interrupt_feedback: Optional[str] = Field(
         None, description="Interrupt feedback from the user on the plan"
     )
+
+
+class TTSRequest(BaseModel):
+    text: str = Field(..., description="The text to convert to speech")
+    voice_type: Optional[str] = Field(
+        "BV700_V2_streaming", description="The voice type to use"
+    )
+    encoding: Optional[str] = Field("mp3", description="The audio encoding format")
+    speed_ratio: Optional[float] = Field(1.0, description="Speech speed ratio")
+    volume_ratio: Optional[float] = Field(1.0, description="Speech volume ratio")
+    pitch_ratio: Optional[float] = Field(1.0, description="Speech pitch ratio")
+    text_type: Optional[str] = Field("plain", description="Text type (plain or ssml)")
+    with_frontend: Optional[int] = Field(
+        1, description="Whether to use frontend processing"
+    )
+    frontend_type: Optional[str] = Field("unitTson", description="Frontend type")

src/tools/__init__.py

@@ -1,6 +1,8 @@
 # Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
 # SPDX-License-Identifier: MIT
 
+import os
+
 from .crawl import crawl_tool
 from .python_repl import python_repl_tool
 from .search import (
@@ -9,6 +11,7 @@ from .search import (
     brave_search_tool,
     arxiv_search_tool,
 )
+from .tts import VolcengineTTS
 from src.config import SELECTED_SEARCH_ENGINE, SearchEngine
 
 # Map search engine names to their respective tools
@@ -25,4 +28,5 @@ __all__ = [
     "crawl_tool",
     "web_search_tool",
     "python_repl_tool",
+    "VolcengineTTS",
 ]

src/tools/tts.py

@@ -0,0 +1,131 @@
+# Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
+# SPDX-License-Identifier: MIT
+
+"""
+Text-to-Speech module using volcengine TTS API.
+"""
+
+import json
+import uuid
+import logging
+import requests
+from typing import Optional, Dict, Any
+
+logger = logging.getLogger(__name__)
+
+
+class VolcengineTTS:
+    """
+    Client for volcengine Text-to-Speech API.
+    """
+
+    def __init__(
+        self,
+        appid: str,
+        access_token: str,
+        cluster: str = "volcano_tts",
+        voice_type: str = "BV700_V2_streaming",
+        host: str = "openspeech.bytedance.com",
+    ):
+        """
+        Initialize the volcengine TTS client.
+
+        Args:
+            appid: Platform application ID
+            access_token: Access token for authentication
+            cluster: TTS cluster name
+            voice_type: Voice type to use
+            host: API host
+        """
+        self.appid = appid
+        self.access_token = access_token
+        self.cluster = cluster
+        self.voice_type = voice_type
+        self.host = host
+        self.api_url = f"https://{host}/api/v1/tts"
+        self.header = {"Authorization": f"Bearer;{access_token}"}
+
+    def text_to_speech(
+        self,
+        text: str,
+        encoding: str = "mp3",
+        speed_ratio: float = 1.0,
+        volume_ratio: float = 1.0,
+        pitch_ratio: float = 1.0,
+        text_type: str = "plain",
+        with_frontend: int = 1,
+        frontend_type: str = "unitTson",
+        uid: Optional[str] = None,
+    ) -> Dict[str, Any]:
+        """
+        Convert text to speech using volcengine TTS API.
+
+        Args:
+            text: Text to convert to speech
+            encoding: Audio encoding format
+            speed_ratio: Speech speed ratio
+            volume_ratio: Speech volume ratio
+            pitch_ratio: Speech pitch ratio
+            text_type: Text type (plain or ssml)
+            with_frontend: Whether to use frontend processing
+            frontend_type: Frontend type
+            uid: User ID (generated if not provided)
+
+        Returns:
+            Dictionary containing the API response and base64-encoded audio data
+        """
+        if not uid:
+            uid = str(uuid.uuid4())
+
+        request_json = {
+            "app": {
+                "appid": self.appid,
+                "token": self.access_token,
+                "cluster": self.cluster,
+            },
+            "user": {"uid": uid},
+            "audio": {
+                "voice_type": self.voice_type,
+                "encoding": encoding,
+                "speed_ratio": speed_ratio,
+                "volume_ratio": volume_ratio,
+                "pitch_ratio": pitch_ratio,
+            },
+            "request": {
+                "reqid": str(uuid.uuid4()),
+                "text": text,
+                "text_type": text_type,
+                "operation": "query",
+                "with_frontend": with_frontend,
+                "frontend_type": frontend_type,
+            },
+        }
+
+        try:
+            logger.debug(f"Sending TTS request for text: {text[:50]}...")
+            response = requests.post(
+                self.api_url, json.dumps(request_json), headers=self.header
+            )
+            response_json = response.json()
+
+            if response.status_code != 200:
+                logger.error(f"TTS API error: {response_json}")
+                return {"success": False, "error": response_json, "audio_data": None}
+
+            if "data" not in response_json:
+                logger.error(f"TTS API returned no data: {response_json}")
+                return {
+                    "success": False,
+                    "error": "No audio data returned",
+                    "audio_data": None,
+                }
+
+            return {
+                "success": True,
+                "response": response_json,
+                "audio_data": response_json["data"],  # Base64 encoded audio data
+            }
+
+        except Exception as e:
+            logger.exception(f"Error in TTS API call: {str(e)}")
+            return {"success": False, "error": str(e), "audio_data": None}