From e9705d18e61212b133835c36d281dc81181d0058 Mon Sep 17 00:00:00 2001 From: Joaquin Madrid Belando Date: Mon, 9 Mar 2026 20:37:19 +0100 Subject: [PATCH] Add TTS_VOLUME parameter (tts_engine.py): adjustable dB volume for TTS audio --- tts_engine.py | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/tts_engine.py b/tts_engine.py index f8d83a8..32ff1de 100644 --- a/tts_engine.py +++ b/tts_engine.py @@ -92,14 +92,15 @@ def _generate_tts_audio(text, lang, mp3_path): return False -def _convert_to_wav(mp3_path, wav_path): +def _convert_to_wav(mp3_path, wav_path, volume_db=0): + _cmd = ['ffmpeg', '-y', '-i', mp3_path, + '-ar', '8000', '-ac', '1', '-sample_fmt', 's16'] + if volume_db != 0: + _cmd += ['-af', 'volume={}dB'.format(volume_db)] + logger.info('(TTS) Aplicando ajuste de volumen: %ddB', volume_db) + _cmd += ['-f', 'wav', wav_path] try: - result = subprocess.run( - ['ffmpeg', '-y', '-i', mp3_path, - '-ar', '8000', '-ac', '1', '-sample_fmt', 's16', - '-f', 'wav', wav_path], - capture_output=True, timeout=60 - ) + result = subprocess.run(_cmd, capture_output=True, timeout=60) if result.returncode != 0: logger.error('(TTS) Error ffmpeg: %s', result.stderr.decode('utf-8', errors='ignore')[:500]) return False @@ -297,7 +298,7 @@ def _encode_ambe_ambeserver(wav_path, ambe_path, host, port): return True -def text_to_ambe(txt_path, ambe_path, language, vocoder_cmd, ambeserver_host='', ambeserver_port=2460): +def text_to_ambe(txt_path, ambe_path, language, vocoder_cmd, ambeserver_host='', ambeserver_port=2460, volume_db=0): if not os.path.isfile(txt_path): logger.warning('(TTS) Archivo de texto no encontrado: %s', txt_path) return False @@ -331,7 +332,7 @@ def text_to_ambe(txt_path, ambe_path, language, vocoder_cmd, ambeserver_host='', if not _generate_tts_audio(text, _tts_lang, _mp3_path): return False - if not _convert_to_wav(_mp3_path, _wav_path): + if not _convert_to_wav(_mp3_path, _wav_path, volume_db): _cleanup([_mp3_path]) return False @@ -383,6 +384,7 @@ def ensure_tts_ambe(config, tts_num): _vocoder_cmd = config['GLOBAL'].get('TTS_VOCODER_CMD', '') _ambeserver_host = config['GLOBAL'].get('TTS_AMBESERVER_HOST', '') _ambeserver_port = config['GLOBAL'].get('TTS_AMBESERVER_PORT', 2460) + _volume_db = config['GLOBAL'].get('TTS_VOLUME', -3) _txt_path = './Audio/{}/ondemand/{}.txt'.format(_lang, _file) _ambe_path = './Audio/{}/ondemand/{}.ambe'.format(_lang, _file) @@ -401,7 +403,7 @@ def ensure_tts_ambe(config, tts_num): logger.warning('(TTS-%d) Archivo de texto no encontrado: %s', tts_num, _txt_path) return None - if text_to_ambe(_txt_path, _ambe_path, _lang, _vocoder_cmd, _ambeserver_host, _ambeserver_port): + if text_to_ambe(_txt_path, _ambe_path, _lang, _vocoder_cmd, _ambeserver_host, _ambeserver_port, _volume_db): return _ambe_path else: if os.path.isfile(_ambe_path):