50 lines
1 KiB
TOML
50 lines
1 KiB
TOML
[package]
|
|
name = "stt-stream"
|
|
version = "0.1.0"
|
|
edition = "2021"
|
|
description = "Local speech-to-text streaming CLI for Fcitx5 integration"
|
|
license = "MIT"
|
|
|
|
[dependencies]
|
|
# Audio capture
|
|
cpal = "0.15"
|
|
# Resampling (48k -> 16k)
|
|
rubato = "0.15"
|
|
# Whisper inference
|
|
whisper-rs = "0.12"
|
|
# Voice activity detection
|
|
# Using silero via ONNX (reserved for future use)
|
|
# ort = { version = "2.0.0-rc.9", default-features = false, features = ["load-dynamic"] }
|
|
# ndarray = "0.16"
|
|
|
|
# Async runtime
|
|
tokio = { version = "1", features = ["full"] }
|
|
|
|
# CLI
|
|
clap = { version = "4", features = ["derive"] }
|
|
|
|
# Serialization for IPC protocol
|
|
serde = { version = "1", features = ["derive"] }
|
|
serde_json = "1"
|
|
|
|
# Error handling
|
|
anyhow = "1"
|
|
thiserror = "1"
|
|
|
|
# Logging
|
|
tracing = "0.1"
|
|
tracing-subscriber = { version = "0.3", features = ["env-filter"] }
|
|
|
|
# Ring buffer for audio (reserved for future use)
|
|
# ringbuf = "0.4"
|
|
|
|
# For downloading models
|
|
hf-hub = "0.3"
|
|
|
|
[features]
|
|
default = []
|
|
cuda = ["whisper-rs/cuda"]
|
|
|
|
[profile.release]
|
|
lto = true
|
|
codegen-units = 1
|