MBROLA(1)

📦 version 3.3, in debian bookworm, debian trixie, ubuntu jammy, ubuntu noble, ubuntu questing, ubuntu resolute, ubuntu stonking

💖 MBROLA 💖

okay bestie, welcome to the General Commands Manual section for mbrola, and honestly? she's iconic. multilingual software speech synthesizer, no cap. let's get into it 🎀

🌸 NAME 🌸

mbrola — a multilingual software speech synthesizer ✨ (yes, multilingual, yes, software, yes, speech, she does it all 💅)

✨ SYNOPSIS ✨

mbrola [options]<voice_database><pho_file>...<output_file>

like, that's the whole vibe. options first, then the voice database, then one or more pho files, then where the output goes. simple. slay. 🦋

💜 DESCRIPTION 💜

so mbrola is a speech synthesizer based on the concatenation of diphones. that's the whole personality. 🧚

she takes a list of phonemes as input, together with prosodic information — that's the duration of phonemes plus a piecewise linear description of pitch — and she produces speech samples on 16 bits (linear), at the sampling frequency of the diphone database. periodt. 💅

⚠️ so it is therefore NOT a Text-To-Speech (TTS) synthesizer, because she does not accept raw text as input. like, she will not read your essay for you bestie. 😭

in order to obtain a full TTS system, you need to use this synthesizer in combination with a text processing system that produces phonetic and prosodic commands. for instance, you may use package freephone to obtain complete English TTS. 🫶

okay and this part is important: a - instead of pho_file or output_file means stdin or stdout. and the extension of output_file (raw, au, wav, aiff) tells the wanted audio format. she's flexible like that. 🌈

👑 OPTIONS 👑

mbrola understands following command line options. (there are a lot. she's thorough. we love a thorough queen. 💖)

  • -h — show summary of options. the little "hi what can i do" moment. ✨
  • -i — display the database information, if any. like a lil ID check for your voice database 🎀
  • -e — ignore fatal errors on unknown diphone. yes, she's being forgiving. yes, that's a choice. 🙈
  • -c comment_char — set COMMENT char (the escape sequence in pho files).
  • -F flush_command — set FLUSH command name.
  • -v volume — volume ratio. a float ratio applied to output samples. 🔊
  • -f freq_ratio — frequency ratio. a float ratio applied to pitch points.
  • -t time_ratio — time ratio. a float ratio applied to phone durations. ⏳
  • -l voice_freq — voice frequency. the target frequency for voice quality, in Hz.
  • -R rename_list — phoneme rename list, of the form: a A b B ...
  • -C clone_list — phoneme clone list, of the form: a A b B ...
  • -I file — initialization file containing one command per line. CLONE, RENAME, VOICE, TIME, FREQ, VOLUME, FLUSH, COMMENT, and IGNORE are available. 👑

🦋 SEE ALSO 🦋

🌸 AUTHOR 🌸

mbrola was written by Dr Thierry Dutoit <dutoit@tcts.fpms.ac.be> 💖

this manual page was written by Igor B. Poretsky <master@goga.energo.ru>, for the Debian project (but may be used by others). 🫶

and that's the whole page bestie!! general commands manual, dated May 17, 2005 — yeah she's an elder icon but she still ate and left no crumbs. mwah 💋✨🎀

📦 where this page comes from