Feature matrix

All 106 backends compiled into the crispasr binary, with their declared capability bits. For an interactive sortable/filterable view, open feature-matrix.html.

Backend	TTS	Voice cloning	Translate	Src/Tgt language	Language detect	Auto-download	Timestamps (native)	Timestamps (CTC)	Word timestamps	Token confidence	Temperature	Beam search	Punctuation toggle	Flash attention	Diarize	GBNF grammar	VAD (internal)	Parallel processors	Beats	Chords	Piano	Pitch	Punctuation Native	S2S	Separate	Streaming	Tab
whisper			✓		✓	✓	✓		✓	✓	✓	✓		✓	✓	✓	✓	✓					✓
nemotron						✓	✓		✓	✓	✓	✓		✓	✓								✓
gigaam						✓	✓		✓	✓				✓									✓
parakeet						✓	✓		✓	✓	✓	✓	✓	✓	✓			✓
reazonspeech						✓	✓		✓	✓	✓	✓	✓	✓	✓			✓
canary			✓	✓		✓	✓	✓	✓	✓	✓	✓	✓	✓	✓			✓
canary-qwen						✓				✓	✓	✓		✓
lfm2-audio	✓					✓						✓												✓
mini-omni2	✓					✓					✓													✓
cohere						✓	✓		✓	✓	✓	✓	✓	✓	✓			✓
granite			✓	✓		✓		✓	✓	✓	✓	✓	✓	✓	✓			✓
granite-4.1			✓	✓		✓		✓	✓	✓	✓	✓	✓	✓	✓			✓
granite-4.1-plus			✓	✓		✓		✓	✓	✓	✓	✓	✓	✓	✓			✓
granite-4.1-nar						✓		✓				✓		✓	✓								✓
voxtral			✓	✓		✓		✓		✓	✓	✓	✓	✓	✓			✓
voxtral4b						✓		✓		✓	✓	✓	✓	✓	✓			✓
voxtral-tts	✓					✓
qwen3			✓	✓		✓		✓		✓	✓	✓	✓	✓	✓			✓
qwen3-1.7b			✓	✓		✓		✓		✓	✓	✓	✓	✓	✓			✓
mega-asr			✓	✓		✓		✓		✓	✓	✓	✓	✓	✓			✓
higgs-stt						✓						✓		✓
fastconformer-ctc						✓		✓		✓					✓			✓
wav2vec2						✓		✓		✓					✓			✓
hubert						✓		✓		✓					✓			✓
data2vec						✓		✓		✓					✓			✓
vibevoice	✓					✓		✓			✓			✓	✓								✓
vibevoice-bitnet	✓					✓		✓			✓			✓	✓								✓
kugelaudio	✓
qwen3-tts	✓	✓		✓		✓					✓			✓												✓
miotts	✓
piano-transcription						✓	✓														✓
moss-tts	✓	✓				✓					✓			✓
moss-tts-local	✓	✓				✓					✓			✓
vibevoice-1.5b	✓	✓				✓		✓			✓			✓	✓								✓
qwen3-tts-customvoice	✓			✓		✓					✓			✓												✓
qwen3-tts-1.7b-base	✓	✓		✓		✓					✓			✓												✓
qwen3-tts-1.7b-customvoice	✓			✓		✓					✓			✓												✓
qwen3-tts-1.7b-voicedesign	✓			✓		✓					✓			✓												✓
orpheus	✓					✓					✓			✓
lex-au-orpheus-de	✓					✓					✓			✓
kartoffel-orpheus-de-natural	✓					✓					✓			✓
kartoffel-orpheus-de-synthetic	✓					✓					✓			✓
chatterbox	✓	✓				✓					✓			✓
chatterbox-turbo	✓	✓				✓					✓			✓
kartoffelbox-turbo	✓	✓				✓					✓			✓
lahgtna-chatterbox	✓	✓				✓					✓			✓
tada	✓					✓					✓
tada-1b	✓					✓					✓
tada-tts-1b	✓					✓					✓
tada-3b-ml	✓					✓					✓
indextts	✓	✓				✓					✓			✓
f5-tts	✓	✓				✓
pocket-tts	✓	✓				✓					✓
fastpitch	✓
kokoro	✓					✓
melotts	✓
piper	✓
outetts	✓	✓				✓					✓			✓
voxcpm2-tts	✓	✓				✓
voxcpm2-vae																								✓
cosyvoice3-tts	✓	✓		✓		✓					✓			✓
m2m100			✓	✓		✓						✓
m2m100-wmt21			✓	✓		✓						✓
madlad			✓	✓		✓						✓
glm-asr			✓	✓		✓		✓		✓	✓	✓	✓	✓	✓
kyutai-stt						✓	✓		✓	✓	✓	✓	✓	✓	✓
firered-asr						✓		✓		✓		✓		✓	✓
moonshine						✓		✓		✓	✓	✓	✓	✓	✓
moonshine-streaming						✓		✓		✓	✓	✓		✓	✓								✓
gemma4-e2b			✓	✓		✓		✓			✓	✓		✓	✓			✓
omniasr						✓		✓		✓	✓	✓		✓	✓
omniasr-300m						✓		✓		✓	✓	✓		✓	✓
omniasr-llm						✓		✓		✓	✓	✓		✓	✓
omniasr-llm-1b						✓		✓		✓	✓	✓		✓	✓
mimo-asr						✓		✓		✓	✓	✓		✓	✓								✓
ark-asr						✓						✓		✓
moss-audio						✓					✓	✓											✓
moss-transcribe						✓						✓											✓
moss-diarize						✓	✓					✓			✓								✓
funasr						✓		✓		✓	✓	✓	✓	✓	✓
fun-asr-mlt-nano						✓		✓		✓	✓	✓	✓	✓	✓
paraformer						✓		✓					✓	✓	✓
sensevoice					✓	✓		✓					✓	✓	✓
sidon																								✓
bark	✓	✓				✓					✓
bark-tts	✓	✓				✓					✓
speecht5	✓					✓
dia	✓					✓					✓
dia-tts	✓					✓					✓
dots-tts	✓					✓					✓
parler-tts	✓					✓					✓
zonos	✓	✓				✓					✓			✓
zonos-tts	✓	✓				✓					✓			✓
csm	✓					✓					✓
csm-tts	✓					✓					✓
sesame	✓					✓					✓
bananamind-tts	✓
omnivoice	✓	✓
omnivoice-singing	✓	✓
htdemucs						✓																			✓
mel-band-roformer						✓																			✓
crepe						✓																✓
btc-chords						✓														✓
tabcnn						✓																					✓
rvc-svc						✓
beat-this						✓													✓

Regenerate with python tools/gen-feature-matrix.py.