src/hooks/useVoice.ts

import { useState, useCallback, useRef, useEffect } from "react";
import { VoiceState } from "@/types/chat";

// Web Speech API types
interface SpeechRecognitionEvent extends Event {
  results: SpeechRecognitionResultList;
}

interface SpeechRecognitionErrorEvent extends Event {
  error: string;
}

interface SpeechRecognitionInstance extends EventTarget {
  continuous: boolean;
  interimResults: boolean;
  lang: string;
  start: () => void;
  stop: () => void;
  abort: () => void;
  onresult: ((event: SpeechRecognitionEvent) => void) | null;
  onerror: ((event: SpeechRecognitionErrorEvent) => void) | null;
  onend: (() => void) | null;
}

declare global {
  interface Window {
    SpeechRecognition: new () => SpeechRecognitionInstance;
    webkitSpeechRecognition: new () => SpeechRecognitionInstance;
  }
}

export function useVoice() {
  const [state, setState] = useState<VoiceState>({
    isListening: false,
    isSpeaking: false,
    transcript: "",
    error: null,
  });

  const recognitionRef = useRef<SpeechRecognitionInstance | null>(null);
  const synthesisRef = useRef<SpeechSynthesisUtterance | null>(null);

  // Initialize speech recognition
  useEffect(() => {
    if (typeof window !== "undefined") {
      const SpeechRecognition =
        window.SpeechRecognition || (window as any).webkitSpeechRecognition;

      if (SpeechRecognition) {
        const recognition = new SpeechRecognition();
        recognition.continuous = false;
        recognition.interimResults = true;
        recognition.lang = "en-US";

        recognition.onresult = (event) => {
          const transcript = Array.from(event.results)
            .map((result) => result[0].transcript)
            .join("");

          setState((prev) => ({ ...prev, transcript }));
        };

        recognition.onerror = (event) => {
          setState((prev) => ({
            ...prev,
            isListening: false,
            error: event.error,
          }));
        };

        recognition.onend = () => {
          setState((prev) => ({ ...prev, isListening: false }));
        };

        recognitionRef.current = recognition;
      }
    }

    return () => {
      if (recognitionRef.current) {
        recognitionRef.current.abort();
      }
    };
  }, []);

  const startListening = useCallback(() => {
    if (!recognitionRef.current) {
      setState((prev) => ({
        ...prev,
        error: "Speech recognition not supported",
      }));
      return;
    }

    setState((prev) => ({
      ...prev,
      isListening: true,
      transcript: "",
      error: null,
    }));

    try {
      recognitionRef.current.start();
    } catch (error) {
      console.error("Failed to start recognition:", error);
    }
  }, []);

  const stopListening = useCallback(() => {
    if (recognitionRef.current) {
      recognitionRef.current.stop();
    }
    setState((prev) => ({ ...prev, isListening: false }));
  }, []);

  const speak = useCallback((text: string) => {
    if (!window.speechSynthesis) {
      setState((prev) => ({ ...prev, error: "Speech synthesis not supported" }));
      return;
    }

    // Cancel any ongoing speech
    window.speechSynthesis.cancel();

    const utterance = new SpeechSynthesisUtterance(text);
    utterance.rate = 1;
    utterance.pitch = 1;
    utterance.volume = 1;

    utterance.onstart = () => {
      setState((prev) => ({ ...prev, isSpeaking: true }));
    };

    utterance.onend = () => {
      setState((prev) => ({ ...prev, isSpeaking: false }));
    };

    utterance.onerror = () => {
      setState((prev) => ({ ...prev, isSpeaking: false }));
    };

    synthesisRef.current = utterance;
    window.speechSynthesis.speak(utterance);
  }, []);

  const stopSpeaking = useCallback(() => {
    window.speechSynthesis.cancel();
    setState((prev) => ({ ...prev, isSpeaking: false }));
  }, []);

  return {
    ...state,
    startListening,
    stopListening,
    speak,
    stopSpeaking,
  };
}