Unverified50% confidenceFactExact time
SteerTTS edge's acoustic model has 9.25M parameters, uses an attention + bidirectional Mamba-3 architecture, and is compressed to int4 through quantization-aware training (QAT).
1
Sources
50%
Confidence
Long-term
Relevance
10/22/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSteerTTS-edge's acoustic model is a 9.25-million-parameter attention + bidirectional Mamba-3 architecture, compressed to int4 precision through quantization-aware training (QAT).91% similarUnverifiedSteerTTS edge's vocoder is a 6.2M-parameter Vocos-style model distilled from charactr/vocos-mel-24khz and subjected to int4 QAT.72% similarUnverifiedSpeech generated by end-to-end TTS models based on deep learning typically achieves a naturalness MOS score of 4.0–4.5 (out of 5), approaching the level of human recordings.67% similarUnverifiedSteerTTS-edge's vocoder is a 6.2-million-parameter Vocos-style model distilled from charactr/vocos-mel-24khz, also using int4 QAT.66% similarVerifiedA complete voice AI interaction involves three serial processing stages: STT, LLM inference, and TTS.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1087447API
curl https://kongchang.com/api/v1/knowledge/claims/1087447MCP
get_claim(id=1087447)