import speech_recognition as sr

# obtain audio from the microphone
r = sr.Recognizer()
with sr.Microphone() as source:
  print("Say something!")
  audio = r.listen(source)

from io import BytesIO

import openai
from speech_recognition.audio import AudioData
from speech_recognition.exceptions import SetupError


def recognize_whisper_api(
  audio_data: "AudioData",
  api_key: str,
  model: str = "whisper-1"
):
  wav_data = BytesIO(audio_data.get_wav_data())
  wav_data.name = "SpeechRecognition_audio.wav"

  transcript = openai.Audio.transcribe(model, wav_data, api_key=api_key, language="ja")
  return transcript["text"]

import axios from 'axios'
import type { AudioQuery } from '../types'

const coeiroinkHost = 'http://127.0.0.1:50031'
const speaker = 0

export const getAudioQuery = async (text: string): Promise<AudioQuery | undefined> => {
 const audioQuery = axios.post(
  `${coeiroinkHost}/audio_query?text=${encodeURIComponent(text)}&speaker=${speaker}`
  , {
   headers: {
    Accept: 'application/json'
   }
  }
 ).then((res) => {
  return res.data as AudioQuery
 }).catch((err) => {
  console.error(err)
  return undefined
 })

 return await audioQuery
}

export const synthesisVoice = async (audioQuery: AudioQuery): Promise<string | undefined> => {
 const voice = await axios.post(
  `${coeiroinkHost}/synthesis?speaker=${speaker}`, audioQuery
  , {
   responseType: 'arraybuffer',
   headers: {
    'Content-Type': 'application/json',
    accept: 'audio/wav'
   }
  }
 ).then((res) => {
  return res.data as string
 }).catch((err) => {
  console.error(err)
  return undefined
 })

 return voice
}

import { PassThrough } from 'stream'
import Speaker from 'speaker'

export const playAudio = (audio: Buffer, sampleRate: number): void => {
 const speaker = new Speaker({
  channels: 1,
  bitDepth: 16,
  sampleRate
 })
 const bufferStream = new PassThrough()
 bufferStream.end(audio)
 bufferStream.pipe(speaker)
}