ComfyUI Node

Fish Audio S2.1 Pro

A ComfyUI node in Runware/Audio/fishaudio with 18 inputs and 1 output.

By Runware·Created 2 years ago·Updated about a month ago· 140
Fish Audio S2.1 Pro
  • speech
  • audioSettings
  • referenceVoices
  • audio
numberResults1
settings.chunkLength300
settings.conditionOnPreviousChunkstrue
settings.earlyStopThreshold1.00
settings.latencynormal
settings.maxTokens1024
settings.minChunkLength50
settings.normalizetrue
settings.normalizeLoudnesstrue
settings.repetitionPenalty1.20
settings.temperature0.70
settings.topP0.70
ttlfalse
ttl_value60
outputFormatMP3
CategoryRunware/Audio/fishaudio

Inputs (18)

NameTypeDefaultDescription
speechRUNWARE_SPEECH
audioSettingsoptRUNWARE_AUDIOSETTINGS
referenceVoicesoptRUNWARE_REFERENCEVOICES
numberResultsoptINT11–4Number of results to generate. Each result uses a different seed, producing variations of the same parameters.
settings.chunkLengthoptINT300100–300Text segment size for processing.
settings.conditionOnPreviousChunksoptBOOLEANtruePrevious audio context for voice consistency across chunks.
settings.earlyStopThresholdoptFLOAT1.000–1Early stopping threshold for batch processing.
settings.latencyoptCOMBOnormalLatency / quality trade-off.
settings.maxTokensoptINT10241–4294967295Maximum audio tokens to generate per text chunk.
settings.minChunkLengthoptINT500–100Minimum characters before splitting into a new chunk.
settings.normalizeoptBOOLEANtrueText normalization for English and Chinese. Improves stability for numbers.
settings.normalizeLoudnessoptBOOLEANtrueNormalize output loudness for consistent perceived volume.
settings.repetitionPenaltyoptFLOAT1.200–2Penalty for repeating audio patterns.
settings.temperatureoptFLOAT0.700–1Expressiveness of the generated speech.
settings.topPoptFLOAT0.700.0001–1Nucleus sampling parameter that controls diversity by limiting the probability mass. Lower values make outputs more focused, higher values increase diversity.
ttloptBOOLEANfalseEnable to set ttl. Off uses the model's default.
ttl_valueoptINT60Time-to-live (TTL) in seconds for generated content. Only applies when `outputType` is `URL`.
outputFormatoptCOMBOMP3File format for the generated audio.

Outputs (1)

NameTypeDescription
audioAUDIO