SpeechBrain

What it does

SpeechBrain is an open-source Python toolkit for building and researching speech and conversational AI systems. Developers and researchers can use pre-built recipes and pretrained models, or define custom deep-learning models, losses, training and evaluation loops, and data pipelines. Its documented task areas include speech recognition, enhancement, separation, text-to-speech, speaker recognition, translation, spoken-language understanding, and audio processing.

The toolkit is installed from PyPI or from its GitHub repository and used in Python projects; the site also describes training recipes configured with YAML files and launched by Python scripts. Work can include processing audio, training or evaluating models, and integrating components into existing pipelines. The stated outputs include tasks such as transcription, speaker verification, enhancement, and source separation. SpeechBrain is software for development rather than a hosted web application; compute and operating-system requirements are not established by the pages. The project is released under Apache License 2.0.

What it can do

• Speech recognition and transcription
• Speech enhancement and source separation
• Text-to-speech and speech-to-speech translation
• Speaker recognition and verification
• Audio feature extraction, augmentation, and sound-event detection
• Train language models and build customizable chatbots for speech pipelines

Works with

• Hugging Face pretrained models

Before you choose

• SpeechBrain is a developer toolkit, not a hosted end-user application; use requires working in a Python environment.
• The official pages do not establish operating-system support, compute requirements, or the runtime cost of individual models and recipes.
• Apache License 2.0 permits redistribution, including commercial use, subject to the license terms.

Listed under

• Coding
• Sandboxes and infrastructure
• Audio processing
• Deep learning
• Developers
• Open-source
• Pytorch toolkit
• Research

Get started

The official site provides PyPI and local installation commands.

1. Install SpeechBrain with `pip install speechbrain`, or clone the GitHub repository and install its requirements and editable package.
2. Choose a documented recipe or pretrained model for the task; review its model or dataset-specific instructions.
3. For a recipe-based training run, use its YAML hyperparameter file with the corresponding Python training script.

Apache-2.0 toolkit; model, dataset, compute and hosting requirements depend on the chosen recipe.

Source: speechbrain.github.io/about.html, checked 1 Nov 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 1 Nov 2026. Each source supports part of the overview; availability and prices can change.

• SpeechBrain: Open-Source Conversational AI for Everyone ↗speechbrain.github.io
• About SpeechBrain ↗speechbrain.github.io
• SpeechBrain: Open-Source Conversational AI for Everyone ↗speechbrain.github.io

Company website

speechbrain.github.io: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 1 Nov 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.