xinyin025/ChatGPT-Next-Web
1
0
Fork 0
mirror of https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web.git synced 2025-10-01 07:36:39 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

feat: add arrayBufferToWav utility for audio processing

Browse Source
This commit is contained in:
Dakai Ou 2024-10-13 17:46:08 +08:00
parent a90090c162
commit 853212cf13
2 changed files with 56 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
app/components/chat.tsx
View File

@ -117,7 +117,7 @@ import { MultimodalContent } from "../client/api";
const localStorage = safeLocalStorage();
import { ClientApi } from "../client/api";
import { createTTSPlayer } from "../utils/audio";
import { createTTSPlayer, arrayBufferToWav } from "../utils/audio";
import { MsEdgeTTS, OUTPUT_FORMAT } from "../utils/ms_edge_tts";
const ttsPlayer = createTTSPlayer();
@ -1228,11 +1228,13 @@ function _Chat() {
}
setSpeechStatus(true);
try {
const waveFile = arrayBufferToWav(audioBuffer);
const audioFile = new Blob([waveFile], { type: "audio/wav" });
const url = uploadAudio(audioFile);
await ttsPlayer.play(audioBuffer, () => {
setSpeechStatus(false);
});
const audioFile = new Blob([audioBuffer], { type: "audio/wav" });
const url = uploadAudio(audioFile);
} catch (e) {
console.error("[OpenAI Speech]", e);
showToast(prettyObject(e));

51
app/utils/audio.ts
View File

@ -43,3 +43,54 @@ export function createTTSPlayer(): TTSPlayer {
return { init, play, stop };
}
export function arrayBufferToWav(buffer: ArrayBuffer): ArrayBuffer {
const numOfChannels = 1; // Mono
const sampleRate = 24000; // 24kHz
const bitsPerSample = 16;
const bytesPerSample = bitsPerSample / 8;
const blockAlign = numOfChannels * bytesPerSample;
const byteRate = sampleRate * blockAlign;
// WAV header size is 44 bytes
const wavHeaderSize = 44;
const dataSize = buffer.byteLength;
const totalSize = wavHeaderSize + dataSize;
const wavBuffer = new ArrayBuffer(totalSize);
const view = new DataView(wavBuffer);
// RIFF chunk descriptor
writeString(view, 0, "RIFF");
view.setUint32(4, totalSize - 8, true); // File size minus RIFF header
writeString(view, 8, "WAVE");
// FMT sub-chunk
writeString(view, 12, "fmt ");
view.setUint32(16, 16, true); // Sub-chunk size (16 for PCM)
view.setUint16(20, 1, true); // Audio format (1 for PCM)
view.setUint16(22, numOfChannels, true); // Number of channels
view.setUint32(24, sampleRate, true); // Sample rate
view.setUint32(28, byteRate, true); // Byte rate
view.setUint16(32, blockAlign, true); // Block align
view.setUint16(34, bitsPerSample, true); // Bits per sample
// Data sub-chunk
writeString(view, 36, "data");
view.setUint32(40, dataSize, true); // Data size
// Write the PCM samples
const audioData = new Uint8Array(buffer);
const wavData = new Uint8Array(wavBuffer);
wavData.set(audioData, wavHeaderSize);
return wavBuffer;
}
// Helper function to write a string to the DataView
function writeString(view: DataView, offset: number, string: string) {
for (let i = 0; i < string.length; i++) {
view.setUint8(offset + i, string.charCodeAt(i));
}
}