Merge pull request #19 from xd009642/feat/proper-audio

Start doing some proper audio handling

Cargo.toml

@@ -25,6 +25,7 @@ opentelemetry = "0.23.0" # otel versions are tied to `init-tracing-opentelemetry
 opentelemetry-otlp = "0.16.0"
 opentelemetry-semantic-conventions = "0.15.0"
 opentelemetry_sdk = { version = "0.23.0", features = ["rt-tokio", "trace"] }
+rubato = "0.15.0"
 serde = { version = "1.0.200", features = ["derive"] }
 serde_json = "1.0.117"
 tokio = { version = "1.37.0", features = ["macros", "signal", "sync", "rt-multi-thread"] }
@@ -37,8 +38,13 @@ tracing-opentelemetry = "0.24.0"
 tracing-subscriber = { version = "0.3.18", features = ["env-filter"] }
 
 [dev-dependencies]
-tracing-test = "0.2.4"
+dasp = { version = "0.11.0", features = ["signal"] }
+hound = "3.5.1"
+tracing-test = { version = "0.2.4", features = ["no-env-filter"] }
 
 [[bin]]
 name = "client"
 required-features = ["tokio-tungstenite", "hound"]
+
+[lints.rust]
+unexpected_cfgs = { level = "warn", check-cfg = ['cfg(tarpaulin_include)'] }

src/api_types.rs

@@ -4,9 +4,25 @@ use serde::{Deserialize, Serialize};
 
 #[derive(Clone, Debug, Eq, PartialEq, Serialize, Deserialize)]
 pub struct StartMessage {
+    /// Trace ID for distributed tracing
     pub trace_id: Option<String>,
+    /// Format information for the audio samples
+    pub format: AudioFormat,
+}
+
+/// Describes the PCM samples coming in. I could have gone for an enum instead of bit_depth +
+/// is_float but I only really plan on doing f32, s16 and ignoring everything else including
+/// compression schemes like mulaw etc. This may change in future.
+#[derive(Copy, Clone, Debug, Eq, PartialEq, Serialize, Deserialize)]
+pub struct AudioFormat {
+    /// Number of channels in the audio
     pub channels: usize,
+    /// Sample rate of the audio
     pub sample_rate: usize,
+    /// Number of bits per sample
+    pub bit_depth: u16,
+    /// Whether audio uses floating point samples
+    pub is_float: bool,
 }
 
 impl Extractor for StartMessage {