Skip to content
SPECIFICTION
Get listed

Microsoft SpeechT5

github.com

Self-hosted
Visit github.com
Spec score 6.9 #209 of 223 in Text-to-Speech Software

Spec sheet

Purpose
SpeechT5 is a unified speech and text pretraining framework for spoken language processing.Read 12 Oct 2026 · github.com
Integrations
The project says its fine-tuned ASR, TTS, and voice-conversion checkpoints can be used with Hugging Face Transformers.Read 12 Oct 2026 · github.com
Support
For help or issues using SpeechT5 models, the README directs users to submit a GitHub issue.Read 12 Oct 2026 · github.com
License
The repository identifies its license as MIT.Read 12 Oct 2026 · github.com
Architecture
The framework combines a shared encoder-decoder network with six speech- and text-specific pre- and post-nets.Read 12 Oct 2026 · github.com
ASR limit
The README states that joint CTC/decoder inference is supported only with batch size 1.Read 12 Oct 2026 · github.com
Commercial use
YesRead 20 Sep 2026 · github.com
Hugging Face
The fine-tuned TTS model is available through Hugging Face Transformers and can be run locally.Read 5 Oct 2026 · huggingface.co
Inference limit
The README states that speech generation is available only with batch size 1.Read 12 Oct 2026 · github.com
Interactive demos
The SpeechT5 instructions link interactive Hugging Face demos for ASR, TTS, and voice conversion.Read 12 Oct 2026 · github.com
Model card limitation
The Hugging Face model card says its language information, risk and limitation details, and several training and evaluation details need more information.Read 5 Oct 2026 · huggingface.co
Model downloads
The repository lists SpeechT5 Base and Large pre-trained models, with model files linked from Hugging Face or Google Drive.Read 5 Oct 2026 · github.com
Models
The repository lists pretrained SpeechT5 Base and Large models, including Base checkpoints with and without LibriSpeech fine-tuning.Read 12 Oct 2026 · github.com
Research audience
The README asks researchers who use the work to cite the SpeechT5 paper.Read 12 Oct 2026 · github.com
Security reporting
Microsoft asks users to report vulnerabilities in Microsoft-owned repositories to MSRC rather than through public GitHub issues.Read 5 Oct 2026 · github.com
Setup
The setup instructions use a Git submodule, editable Fairseq installation, and ESPnet installed with pip.Read 12 Oct 2026 · github.com
Tasks
The project evaluates automatic speech recognition, text-to-speech, speech translation, voice conversion, speech enhancement, and speaker identification.Read 12 Oct 2026 · github.com

Where Specifiction places it

Alternatives to Microsoft SpeechT5All alternatives →