Add endpointing (#54)

python-api-examples/speech-recognition-from-microphone-with-endpoint-detection.py

@@ -0,0 +1,85 @@
+#!/usr/bin/env python3
+
+# Real-time speech recognition from a microphone with sherpa-onnx Python API
+# with endpoint detection.
+#
+# Please refer to
+# https://k2-fsa.github.io/sherpa/onnx/pretrained_models/index.html
+# to download pre-trained models
+
+import sys
+
+try:
+    import sounddevice as sd
+except ImportError as e:
+    print("Please install sounddevice first. You can use")
+    print()
+    print("  pip install sounddevice")
+    print()
+    print("to install it")
+    sys.exit(-1)
+
+import sherpa_onnx
+
+
+def create_recognizer():
+    # Please replace the model files if needed.
+    # See https://k2-fsa.github.io/sherpa/onnx/pretrained_models/index.html
+    # for download links.
+    recognizer = sherpa_onnx.OnlineRecognizer(
+        tokens="./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/tokens.txt",
+        encoder="./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/encoder-epoch-99-avg-1.onnx",
+        decoder="./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/decoder-epoch-99-avg-1.onnx",
+        joiner="./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/joiner-epoch-99-avg-1.onnx",
+        num_threads=4,
+        sample_rate=16000,
+        feature_dim=80,
+        enable_endpoint_detection=True,
+        rule1_min_trailing_silence=2.4,
+        rule2_min_trailing_silence=1.2,
+        rule3_min_utterance_length=300,  # it essentially disables this rule
+    )
+    return recognizer
+
+
+def main():
+    print("Started! Please speak")
+    recognizer = create_recognizer()
+    sample_rate = 16000
+    samples_per_read = int(0.1 * sample_rate)  # 0.1 second = 100 ms
+    last_result = ""
+    stream = recognizer.create_stream()
+
+    last_result = ""
+    segment_id = 0
+    with sd.InputStream(channels=1, dtype="float32", samplerate=sample_rate) as s:
+        while True:
+            samples, _ = s.read(samples_per_read)  # a blocking read
+            samples = samples.reshape(-1)
+            stream.accept_waveform(sample_rate, samples)
+            while recognizer.is_ready(stream):
+                recognizer.decode_stream(stream)
+
+            is_endpoint = recognizer.is_endpoint(stream)
+
+            result = recognizer.get_result(stream)
+
+            if result and (last_result != result):
+                last_result = result
+                print(f"{segment_id}: {result}")
+
+            if result and is_endpoint:
+                segment_id += 1
+                recognizer.reset(stream)
+
+
+if __name__ == "__main__":
+    devices = sd.query_devices()
+    print(devices)
+    default_input_device_idx = sd.default.device[0]
+    print(f'Use default device: {devices[default_input_device_idx]["name"]}')
+
+    try:
+        main()
+    except KeyboardInterrupt:
+        print("\nCaught Ctrl + C. Exiting")