Nodes/ComfyUI-Index-TTS-25/IndexTTS 2 / 2.5 Synthesize
ComfyUI Node

IndexTTS 2 / 2.5 Synthesize

IndexTTS 2/2.5 zero-shot multilingual and emotion-controllable speech synthesis.

By joyfoxai·Created 12 days ago·Updated 5 days ago· 0
IndexTTS 2 / 2.5 Synthesize
  • model
  • speaker_audio
  • emotion_audio
  • audio
text欢迎使用 IndexTTS 2.5。
languageZH
emotion_modesame_as_speaker
emotion_weight0.65
extra_emotion_text
happy0.00
angry0.00
sad0.00
fearful0.00
disgusted0.00
melancholic0.00
surprised0.00
calm0.00
emotion_randomfalse
duration_factor1.00
interval_silence_ms200
text_normalizationtrue
max_text_tokens_per_segment120
do_sampletrue
temperature0.8
top_p0.80
top_k30
num_beams3
repetition_penalty10.0
length_penalty0.0
max_mel_tokens1500
seed0
Categoryaudio/IndexTTS

Inputs (30)

NameTypeDefaultDescription
modelINDEXTTS25_MODEL
speaker_audioAUDIO
textSTRING欢迎使用 IndexTTS 2.5。
languageCOMBOZH5 options: ZH, EN, JA, AR, ES
emotion_modeCOMBOsame_as_speaker5 options: same_as_speaker, reference_audio, emotion_vector, emotion_text, extra_emotion_text
emotion_weightFLOAT0.650–1
extra_emotion_textSTRING
happyFLOAT0.000–1
angryFLOAT0.000–1
sadFLOAT0.000–1
fearfulFLOAT0.000–1
disgustedFLOAT0.000–1
melancholicFLOAT0.000–1
surprisedFLOAT0.000–1
calmFLOAT0.000–1
emotion_randomBOOLEANfalse
duration_factorFLOAT1.000.5–2
interval_silence_msINT2000–5000
text_normalizationBOOLEANtrue
max_text_tokens_per_segmentINT12020–400
do_sampleBOOLEANtrue
temperatureFLOAT0.80.1–2
top_pFLOAT0.800–1
top_kINT300–100
num_beamsINT31–10
repetition_penaltyFLOAT10.00.1–20
length_penaltyFLOAT0.0-2–2
max_mel_tokensINT150050–4096
seedINT00–9223372036854776000
emotion_audiooptAUDIO

Outputs (1)

NameTypeDescription
audioAUDIO