MBROLA

by Mons University / open source

Diphone-based TTS engine — paired with eSpeak NG for more natural output.

TL;DR

Diphone-based TTS engine — paired with eSpeak NG for more natural output.

Best for eSpeak NG users wanting more natural diphone-based voices in 35+ languages. Pricing: free (AGPL since 2018).

Category
Open source
License
—
Stars
—
Last push
—
Pricing
free (AGPL since 2018)
Platforms
Linux, macOS, Windows

What it is

MBROLA is a diphone-based speech synthesizer often paired with eSpeak NG as the voice backend — eSpeak handles phonemization and MBROLA produces audio. Re-released under AGPL in 2018 after years as freeware. Consent posture: synthetic-only.

Best for: eSpeak NG users wanting more natural diphone-based voices in 35+ languages.
Watch out for: Voice databases vary in quality; some languages still classical-sounding.

Install / use

sudo apt install mbrola

Features

Speaker diarizationNo
Word-level timestampsNo
Streaming / real-timeNo
Languages supported35
HIPAA eligibleNo

MBROLA vs Whipscribe

FeatureMBROLAWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfree (AGPL since 2018)free beta
Speaker diarization—Yes
Word timestamps—Yes
Streaming—No
Languages3599
PlatformsLinux, macOS, WindowsWeb, API, MCP

Alternatives to MBROLA

OpenAI Whisper
OpenAI

The reference open-source multilingual ASR model from OpenAI.

OSS · MIT ★ 98.1k
whisper.cpp
Georgi Gerganov

C/C++ port of Whisper — runs on anything, from a Raspberry Pi to Apple Silicon.

OSS · MIT ★ 48.8k
faster-whisper
SYSTRAN

4× faster than reference Whisper using CTranslate2 — production sweet spot.

OSS · MIT ★ 22.3k

Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.