whatmodel.sh / tts / SpeedySpeech (Coqui LJ Speech)
SpeedySpeech (Coqui LJ Speech)
↗ share
+ contribute
SpeedySpeech (Coqui LJ Speech) is an open-weights text-to-speech model from Coqui TTS / Eren Gölge @erogol, released September 2021. Licence: apache 2.0.
Catalogue identity notes
Exact registry entry tts_models/en/ljspeech/speedy-speech. Default vocoder vocoder_models/en/ljspeech/hifigan_v2. Model archive responded HTTP 200 to a HEAD request on 2026-09-29. CPU command documented by maintained Coqui TTS; phoneme models may require the espeak-ng system package. Memory is an estimated 4GB runtime budget, not measured usage. No original-paper MOS transferred to this separately trained checkpoint.
Model information sources
$ how it compares
Scored text-to-speech models released closest to SpeedySpeech (Coqui LJ Speech), out of 148 with a score.
| # | model | score | where |
|---|---|---|---|
| 81 |  Azure Ameha Neural | 50±16 | comparison only |
| 82 | Azure Mekdes Neural | 50±16 | comparison only |
| 124 |  VITS (VCTK multi-speaker) | 34±5 | local |
| 135 | Glow-TTS + HiFi-GAN (VCTK, VITS study) | 28±6 | comparison only |
| 142 | Tacotron 2 + HiFi-GAN (VCTK, VITS study) | 19±6 | comparison only |
see all text-to-speech models · find the ones that fit your machine
$ where to run it
Other ways to run it
- Run it as published · needs ~4 GB · from github.com/idiap
uv run --python 3.11 --with 'coqui-tts[languages]' --with torch --with torchaudio --with torchcodec tts --model_name tts_models/en/ljspeech/speedy-speech --text "Hello world." --out_path speech.wav --device cpu
Weights and documentation: github.com/idiap ↗
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json