mirrors/hermes-agent
1
0
Fork 0
mirror of https://github.com/NousResearch/hermes-agent.git synced 2026-05-08 03:01:47 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 7

fix(whatsapp): auto-convert mp3/wav to ogg/opus in send-media for native voice bubbles

Browse source 
WhatsApp bridge (bridge.js) only sets ptt:true when file extension is .ogg
or .opus, causing mp3/wav files (from Edge TTS, NeuTTS, etc.) to arrive
as file attachments instead of voice bubbles — silently, with no error.

Fix: when audio type is sent with a non-ogg/opus format, run ffmpeg
conversion to ogg/opus in a temp file before sending. This makes
send_voice() self-sufficient regardless of what format the caller provides.

Fallback: if ffmpeg is unavailable, original buffer is sent (previous
behaviour) with a console.warn — no crash.

Addresses veloguardian's review comment on PR #4992.
This commit is contained in:
Bartok 2026-04-21 16:29:27 -04:00 • committed by Teknium
parent 45949e944a
commit dad62c4c47
1 changed files with 28 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

31
scripts/whatsapp-bridge/bridge.js
View file

@ -23,8 +23,10 @@ import express from 'express';
import { Boom } from '@hapi/boom';
import pino from 'pino';
import path from 'path';
import { mkdirSync, readFileSync, writeFileSync, existsSync, readdirSync } from 'fs';
import { mkdirSync, readFileSync, writeFileSync, existsSync, readdirSync, unlinkSync } from 'fs';
import { randomBytes } from 'crypto';
import { execSync } from 'child_process';
import { tmpdir } from 'os';
import qrcode from 'qrcode-terminal';
import { matchesAllowedUser, parseAllowedUsers } from './allowlist.js';
@ -505,8 +507,31 @@ app.post('/send-media', async (req, res) => {
msgPayload = { video: buffer, caption: caption || undefined, mimetype: MIME_MAP[ext] || 'video/mp4' };
break;
case 'audio': {
const audioMime = (ext === 'ogg' || ext === 'opus') ? 'audio/ogg; codecs=opus' : 'audio/mpeg';
msgPayload = { audio: buffer, mimetype: audioMime, ptt: ext === 'ogg' || ext === 'opus' };
// WhatsApp only renders a native voice bubble (ptt) when the file is ogg/opus.
// If the caller passes mp3, wav, m4a etc. (e.g. from Edge TTS / NeuTTS),
// silently convert to ogg/opus via ffmpeg so ptt is always honoured.
let audioBuffer = buffer;
let audioExt = ext;
const needsConversion = !['ogg', 'opus'].includes(ext);
let tmpPath = null;
if (needsConversion) {
tmpPath = path.join(tmpdir(), `hermes_voice_${randomBytes(6).toString('hex')}.ogg`);
try {
execSync(
`ffmpeg -y -i ${JSON.stringify(filePath)} -ar 48000 -ac 1 -c:a libopus ${JSON.stringify(tmpPath)}`,
{ timeout: 30000, stdio: 'pipe' }
);
audioBuffer = readFileSync(tmpPath);
audioExt = 'ogg';
} catch (convErr) {
// ffmpeg not available or conversion failed — fall back to original format
console.warn('[bridge] ffmpeg conversion failed, sending as file attachment:', convErr.message);
} finally {
try { if (tmpPath && existsSync(tmpPath)) unlinkSync(tmpPath); } catch (_) {}
}
}
const audioMime = (audioExt === 'ogg' || audioExt === 'opus') ? 'audio/ogg; codecs=opus' : 'audio/mpeg';
msgPayload = { audio: audioBuffer, mimetype: audioMime, ptt: audioExt === 'ogg' || audioExt === 'opus' };
break;
}
case 'document':