asr.ts

import spokestackService, { SpokestackASRConfig } from './spokestackASRService'

import { SpeechClient } from '@google-cloud/speech'

/**
 * A one-off method for processing speech to text
 * using Spokestack ASR.
 *
 *
 * ```js
 * import fileUpload from 'express-fileupload'
 * import { asr } from 'spokestack'
 * import express from 'express'
 *
 * const expressApp = express()
 *
 * expressApp.post('/asr', fileUpload(), (req, res) => {
 *   const sampleRate = Number(req.body.sampleRate)
 *   const audio = req.files.audio
 *   if (isNaN(sampleRate)) {
 *     res.status(400)
 *     res.send('Parameter required: "sampleRate"')
 *     return
 *   }
 *   if (!audio) {
 *     res.status(400)
 *     res.send('Parameter required: "audio"')
 *     return
 *   }
 *   asr(Buffer.from(audio.data.buffer), sampleRate)
 *     .then((text) => {
 *       res.status(200)
 *       res.json({ text })
 *     })
 *     .catch((error: Error) => {
 *       console.error(error)
 *       res.status(500)
 *       res.send('Unknown error during speech recognition. Check server logs.')
 *     })
 * })
 *
 * ```
 */
export function asr(
  content: string | Uint8Array,
  config: SpokestackASRConfig
): Promise<string | null> {
  return new Promise((resolve, reject) => {
    spokestackService(config, (response) => {
      // console.log('[asr] response', response)
      if (response.status === 'ok' && response.final) {
        resolve(
          response.hypotheses
            .map((value) => value && value.transcript)
            .filter(Boolean)
            .join('\n')
        )
      } else if (response.status === 'error') {
        console.error('Unexpected ASR error', response.error)
        reject(new Error(response.error))
      }
    })
      .then((spokestackSocket) => {
        spokestackSocket.on('error', reject)
        spokestackSocket.send(content)
        // Send an empty buffer to signal that the transaction is done
        spokestackSocket.send(Buffer.from(''))
      })
      .catch(reject)
  })
}

/**
 * A one-off method for processing speech to text
 * using Google Speech.
 *
 *
 * ```js
 * import fileUpload from 'express-fileupload'
 * import { googleASR } from 'spokestack'
 * import express from 'express'
 *
 * const expressApp = express()
 *
 * expressApp.post('/asr', fileUpload(), (req, res) => {
 *   const sampleRate = Number(req.body.sampleRate)
 *   const audio = req.files.audio
 *   if (isNaN(sampleRate)) {
 *     res.status(400)
 *     res.send('Parameter required: "sampleRate"')
 *     return
 *   }
 *   if (!audio) {
 *     res.status(400)
 *     res.send('Parameter required: "audio"')
 *     return
 *   }
 *   googleASR(Buffer.from(audio.data.buffer), sampleRate)
 *     .then((text) => {
 *       res.status(200)
 *       res.json({ text })
 *     })
 *     .catch((error: Error) => {
 *       console.error(error)
 *       res.status(500)
 *       res.send('Unknown error during speech recognition. Check server logs.')
 *     })
 * })
 *
 * ```
 */
export async function googleASR(
  content: string | Uint8Array,
  sampleRate: number
): Promise<string | null> {
  const client = new SpeechClient()
  const fullResponse = await client.recognize({
    audio: { content },
    config: {
      sampleRateHertz: sampleRate,
      encoding: 'LINEAR16',
      languageCode: 'en-US'
    }
  })
  // console.log(fullResponse)
  const response = fullResponse[0]
  if (!response || !response.results || !response.results.length) {
    return null
  }
  return response.results
    .map((result) => (result.alternatives ? result.alternatives[0].transcript : undefined))
    .filter(Boolean)
    .join('\n')
}