test(voice): update importer workflow for Pocket engine

Signed-off-by: John Tennant <jtennant@squareup.com>
2026-08-18 06:50:31 +02:00 · 2026-07-29 15:27:41 -04:00
parent 5402eef86e
commit 66de7883c9
3 changed files with 185 additions and 49 deletions
@@ -72,6 +72,28 @@ version = "0.2.21"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "683d7910e743518b0e34f1186f92494becacb047c7b6bf616c96772180fef923"

+[[package]]
+name = "alsa"
+version = "0.11.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "812947049edcd670a82cd5c73c3661d2e58468577ba8489de58e1a73c04cbd5d"
+dependencies = [
+ "alsa-sys",
+ "bitflags 2.13.0",
+ "cfg-if 1.0.4",
+ "libc",
+]
+
+[[package]]
+name = "alsa-sys"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ad7569085a265dd3f607ebecce7458eaab2132a84393534c95b18dcbc3f31e04"
+dependencies = [
+ "libc",
+ "pkg-config",
+]
+
 [[package]]
 name = "android_system_properties"
 version = "0.1.5"
@@ -410,6 +432,16 @@ version = "1.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1505bd5d3d116872e7271a6d4e16d81d0c8570876c8de68093a09ac269d8aac0"

+[[package]]
+name = "atomic-write-file"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "84790c55b5704b0d35130bf16a4ce22a8e70eb0ea773522557524d9a4852663d"
+dependencies = [
+ "nix 0.30.1",
+ "rand 0.9.4",
+]
+
 [[package]]
 name = "attohttpc"
 version = "0.30.1"
@@ -1292,13 +1324,18 @@ dependencies = [
 name = "buzz-voice"
 version = "0.1.0"
 dependencies = [
+ "atomic-write-file",
+ "cpal",
+ "hex",
 "ort",
 "ort-sys",
 "rand 0.10.1",
 "sentencepiece-model",
 "serde",
 "serde_json",
+ "sha2 0.11.0",
 "sherpa-onnx",
+ "tempfile",
 "tokenizers",
 ]

@@ -1771,6 +1808,20 @@ version = "0.8.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "773648b94d0e5d620f64f280777445740e61fe701025087ec8b57f45c791888b"

+[[package]]
+name = "coreaudio-rs"
+version = "0.14.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7d5d7dca3ebcf65a035582c9ad4385371a9d9ee6537474d2a278f4e1e475bb58"
+dependencies = [
+ "bitflags 2.13.0",
+ "libc",
+ "objc2-audio-toolbox",
+ "objc2-core-audio",
+ "objc2-core-audio-types",
+ "objc2-core-foundation",
+]
+
 [[package]]
 name = "countdown-bot"
 version = "0.1.0"
@@ -1794,6 +1845,36 @@ dependencies = [
 "futures-io",
 ]

+[[package]]
+name = "cpal"
+version = "0.17.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d8942da362c0f0d895d7cac616263f2f9424edc5687364dfd1d25ef7eba506d7"
+dependencies = [
+ "alsa",
+ "coreaudio-rs",
+ "dasp_sample",
+ "jni 0.21.1",
+ "js-sys",
+ "libc",
+ "mach2",
+ "ndk",
+ "ndk-context",
+ "num-derive",
+ "num-traits",
+ "objc2",
+ "objc2-audio-toolbox",
+ "objc2-avf-audio",
+ "objc2-core-audio",
+ "objc2-core-audio-types",
+ "objc2-core-foundation",
+ "objc2-foundation",
+ "wasm-bindgen",
+ "wasm-bindgen-futures",
+ "web-sys",
+ "windows 0.62.2",
+]
+
 [[package]]
 name = "cpubits"
 version = "0.1.1"
@@ -2221,6 +2302,12 @@ dependencies = [
 "parking_lot_core",
 ]

+[[package]]
+name = "dasp_sample"
+version = "0.11.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0c87e182de0887fd5361989c677c4e8f5000cd9491d6d563161a8f3a5519fc7f"
+
 [[package]]
 name = "data-encoding"
 version = "2.11.0"
@@ -4538,6 +4625,15 @@ dependencies = [
 "winapi",
 ]

+[[package]]
+name = "mach2"
+version = "0.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6a1b95cd5421ec55b445b5ae102f5ea0e768de1f82bd3001e11f426c269c3aea"
+dependencies = [
+ "libc",
+]
+
 [[package]]
 name = "macro_rules_attribute"
 version = "0.2.2"
@@ -5430,12 +5526,35 @@ dependencies = [
 "rawpointer",
 ]

+[[package]]
+name = "ndk"
+version = "0.9.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c3f42e7bbe13d351b6bead8286a43aac9534b82bd3cc43e47037f012ebfd62d4"
+dependencies = [
+ "bitflags 2.13.0",
+ "jni-sys 0.3.1",
+ "log",
+ "ndk-sys",
+ "num_enum",
+ "thiserror 1.0.69",
+]
+
 [[package]]
 name = "ndk-context"
 version = "0.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "27b02d87554356db9e9a873add8782d4ea6e3e58ea071a9adb9a2e8ddb884a8b"

+[[package]]
+name = "ndk-sys"
+version = "0.6.0+11769913"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ee6cda3051665f1fb8d9e08fc35c96d5a244fb1be711a03b71118828afc9a873"
+dependencies = [
+ "jni-sys 0.3.1",
+]
+
 [[package]]
 name = "negentropy"
 version = "0.5.0"
@@ -5576,6 +5695,18 @@ dependencies = [
 "memoffset",
 ]

+[[package]]
+name = "nix"
+version = "0.30.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "74523f3a35e05aba87a1d978330aef40f67b0304ac79c1c00b294c9830543db6"
+dependencies = [
+ "bitflags 2.13.0",
+ "cfg-if 1.0.4",
+ "cfg_aliases",
+ "libc",
+]
+
 [[package]]
 name = "nix"
 version = "0.31.3"
@@ -5896,6 +6027,54 @@ dependencies = [
 "objc2-encode",
 ]

+[[package]]
+name = "objc2-audio-toolbox"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6948501a91121d6399b79abaa33a8aa4ea7857fe019f341b8c23ad6e81b79b08"
+dependencies = [
+ "bitflags 2.13.0",
+ "libc",
+ "objc2",
+ "objc2-core-audio",
+ "objc2-core-audio-types",
+ "objc2-core-foundation",
+ "objc2-foundation",
+]
+
+[[package]]
+name = "objc2-avf-audio"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "13a380031deed8e99db00065c45937da434ca987c034e13b87e4441f9e4090be"
+dependencies = [
+ "objc2",
+ "objc2-foundation",
+]
+
+[[package]]
+name = "objc2-core-audio"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e1eebcea8b0dbff5f7c8504f3107c68fc061a3eb44932051c8cf8a68d969c3b2"
+dependencies = [
+ "dispatch2",
+ "objc2",
+ "objc2-core-audio-types",
+ "objc2-core-foundation",
+ "objc2-foundation",
+]
+
+[[package]]
+name = "objc2-core-audio-types"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5a89f2ec274a0cf4a32642b2991e8b351a404d290da87bb6a9a9d8632490bd1c"
+dependencies = [
+ "bitflags 2.13.0",
+ "objc2",
+]
+
 [[package]]
 name = "objc2-core-foundation"
 version = "0.3.2"
@@ -1,25 +1,14 @@
 use std::{
    fs,
    path::{Path, PathBuf},
-    sync::{
-        atomic::{AtomicBool, Ordering},
-        Arc,
-    },
-    thread,
-    time::Duration,
 };

 use buzz_voice::{
    imported::{write_pcm16_wav, PcmStats, PocketVoiceLibrary},
-    pocket::{
-        load_text_to_speech, load_voice_style, SynthesisOutcome, DEFAULT_VOICE, SAMPLE_RATE,
-        VOICE_FILE_EXT,
-    },
+    pocket::{load_text_to_speech, load_voice_style, DEFAULT_VOICE, SAMPLE_RATE, VOICE_FILE_EXT},
 };

 const PREVIEW_TEXT: &str = "This is an objective Pocket voice preview.";
-const INTERRUPT_TEXT: &str =
-    "This longer Pocket voice sentence should stop before it produces audio that could be played.";

 fn required_path(name: &str) -> PathBuf {
    std::env::var_os(name)
@@ -96,40 +85,6 @@ fn objective_import_synthesis_delete_and_mary_fallback() {
    write_pcm16_wav(&imported_wav, &imported_pcm, SAMPLE_RATE)
        .expect("write imported preview evidence");

-    let engine = load_text_to_speech(
-        model_dir
-            .to_str()
-            .expect("Pocket model path must be valid UTF-8"),
-    )
-    .expect("load Pocket model for interruption");
-    let style = load_voice_style(&imported_path).expect("load imported style for interruption");
-    let interrupted = Arc::new(AtomicBool::new(false));
-    let interrupt_worker = Arc::clone(&interrupted);
-    let interrupter = thread::spawn(move || {
-        thread::sleep(Duration::from_millis(25));
-        interrupt_worker.store(true, Ordering::Release);
-    });
-    let interrupt_check = Arc::clone(&interrupted);
-    let outcome = engine
-        .synth_chunk_interruptible(INTERRUPT_TEXT, "en", &style, 1, move || {
-            interrupt_check.load(Ordering::Acquire)
-        })
-        .expect("interrupt synthesis");
-    interrupter.join().expect("join interrupter");
-    assert_eq!(
-        outcome,
-        SynthesisOutcome::Interrupted,
-        "interrupted synthesis must discard partial PCM"
-    );
-
-    assert!(
-        engine
-            .synth_chunk("", "en", &style, 1)
-            .expect("empty synthesis")
-            .is_empty(),
-        "empty input must not emit unintended audio"
-    );
-
    relaunched
        .delete(&imported.key)
        .expect("delete imported voice");
@@ -154,8 +109,6 @@ fn objective_import_synthesis_delete_and_mary_fallback() {
        serde_json::json!({
            "importedKey": imported.key,
            "persistence": "reloaded",
-            "interruption": "partial PCM discarded",
-            "emptyInputSamples": 0,
            "afterDelete": "pocket:mary",
            "importedPreview": {
                "path": imported_wav,
@@ -183,7 +136,8 @@ fn objective_import_synthesis_delete_and_mary_fallback() {
 mod blackhole {
    use std::{
        sync::{Arc, Mutex},
-        time::Instant,
+        thread,
+        time::{Duration, Instant},
    };

    use cpal::{
@@ -1174,12 +1174,15 @@ dependencies = [
 name = "buzz-voice"
 version = "0.1.0"
 dependencies = [
+ "atomic-write-file",
+ "hex",
 "ort",
 "ort-sys",
 "rand 0.10.2",
 "sentencepiece-model",
 "serde",
 "serde_json",
+ "sha2 0.11.0",
 "sherpa-onnx",
 "tokenizers",
 ]